{
    "segments": [
        {
            "id": 1,
            "seek": 3000,
            "start": 0.0,
            "end": 3.7,
            "text": " Hello, I am Miguel Guevara.",
            "tokens": [
                50414,
                2425,
                11,
                286,
                669,
                29150,
                44847,
                85,
                2419,
                13,
                50514
            ],
            "temperature": 0.0,
            "avg_logprob": -0.28414948453608246,
            "compression_ratio": 1.6422413793103448,
            "no_speech_prob": 0.4638671875,
            "words": [
                {
                    "start": 0.0,
                    "end": 0.4,
                    "word": " Hello",
                    "probability": 0.031280517578125
                },
                {
                    "start": 0.4,
                    "end": 1.94,
                    "word": ",",
                    "probability": 0.59130859375
                },
                {
                    "start": 1.94,
                    "end": 2.22,
                    "word": " I",
                    "probability": 0.96484375
                },
                {
                    "start": 2.22,
                    "end": 2.44,
                    "word": " am",
                    "probability": 0.75390625
                },
                {
                    "start": 2.44,
                    "end": 2.84,
                    "word": " Miguel",
                    "probability": 0.751953125
                },
                {
                    "start": 2.84,
                    "end": 3.2,
                    "word": " Guevara",
                    "probability": 0.84033203125
                },
                {
                    "start": 3.2,
                    "end": 3.7,
                    "word": ".",
                    "probability": 0.5751953125
                }
            ]
        },
        {
            "id": 2,
            "seek": 3000,
            "start": 3.8,
            "end": 9.92,
            "text": " And I'm Brian Gibson. I work in privacy infrastructure at Google, in particular on differential privacy.",
            "tokens": [
                50564,
                400,
                286,
                478,
                10765,
                42250,
                13,
                286,
                589,
                294,
                11427,
                6896,
                412,
                3329,
                11,
                294,
                1729,
                322,
                15756,
                11427,
                13,
                50814
            ],
            "temperature": 0.0,
            "avg_logprob": -0.28414948453608246,
            "compression_ratio": 1.6422413793103448,
            "no_speech_prob": 0.4638671875,
            "words": [
                {
                    "start": 3.8,
                    "end": 4.36,
                    "word": " And",
                    "probability": 0.79296875
                },
                {
                    "start": 4.36,
                    "end": 4.62,
                    "word": " I'm",
                    "probability": 0.912841796875
                },
                {
                    "start": 4.62,
                    "end": 4.82,
                    "word": " Brian",
                    "probability": 0.60595703125
                },
                {
                    "start": 4.82,
                    "end": 5.1,
                    "word": " Gibson",
                    "probability": 0.8349609375
                },
                {
                    "start": 5.1,
                    "end": 5.5,
                    "word": ".",
                    "probability": 0.80908203125
                },
                {
                    "start": 5.5,
                    "end": 5.76,
                    "word": " I",
                    "probability": 0.98681640625
                },
                {
                    "start": 5.76,
                    "end": 6.12,
                    "word": " work",
                    "probability": 0.814453125
                },
                {
                    "start": 6.12,
                    "end": 6.32,
                    "word": " in",
                    "probability": 0.89794921875
                },
                {
                    "start": 6.32,
                    "end": 6.6,
                    "word": " privacy",
                    "probability": 0.6279296875
                },
                {
                    "start": 6.6,
                    "end": 7.16,
                    "word": " infrastructure",
                    "probability": 0.8740234375
                },
                {
                    "start": 7.16,
                    "end": 7.4,
                    "word": " at",
                    "probability": 0.8955078125
                },
                {
                    "start": 7.4,
                    "end": 7.62,
                    "word": " Google",
                    "probability": 0.8603515625
                },
                {
                    "start": 7.62,
                    "end": 7.82,
                    "word": ",",
                    "probability": 0.70654296875
                },
                {
                    "start": 7.82,
                    "end": 7.92,
                    "word": " in",
                    "probability": 0.90234375
                },
                {
                    "start": 7.92,
                    "end": 8.28,
                    "word": " particular",
                    "probability": 0.83251953125
                },
                {
                    "start": 8.28,
                    "end": 8.54,
                    "word": " on",
                    "probability": 0.72216796875
                },
                {
                    "start": 8.54,
                    "end": 8.86,
                    "word": " differential",
                    "probability": 0.70458984375
                },
                {
                    "start": 8.86,
                    "end": 9.36,
                    "word": " privacy",
                    "probability": 0.83154296875
                },
                {
                    "start": 9.36,
                    "end": 9.92,
                    "word": ".",
                    "probability": 0.962890625
                }
            ]
        },
        {
            "id": 3,
            "seek": 3000,
            "start": 10.08,
            "end": 15.06,
            "text": " And we will talk to you about today our differential privacy.",
            "tokens": [
                50864,
                400,
                321,
                486,
                751,
                281,
                291,
                466,
                965,
                527,
                15756,
                11427,
                13,
                51064
            ],
            "temperature": 0.0,
            "avg_logprob": -0.28414948453608246,
            "compression_ratio": 1.6422413793103448,
            "no_speech_prob": 0.4638671875,
            "words": [
                {
                    "start": 10.08,
                    "end": 10.54,
                    "word": " And",
                    "probability": 0.8486328125
                },
                {
                    "start": 10.54,
                    "end": 10.76,
                    "word": " we",
                    "probability": 0.8916015625
                },
                {
                    "start": 10.76,
                    "end": 10.94,
                    "word": " will",
                    "probability": 0.7626953125
                },
                {
                    "start": 10.94,
                    "end": 11.2,
                    "word": " talk",
                    "probability": 0.81396484375
                },
                {
                    "start": 11.2,
                    "end": 11.34,
                    "word": " to",
                    "probability": 0.9013671875
                },
                {
                    "start": 11.34,
                    "end": 11.54,
                    "word": " you",
                    "probability": 0.8935546875
                },
                {
                    "start": 11.54,
                    "end": 11.8,
                    "word": " about",
                    "probability": 0.7939453125
                },
                {
                    "start": 11.8,
                    "end": 12.22,
                    "word": " today",
                    "probability": 0.771484375
                },
                {
                    "start": 12.22,
                    "end": 13.6,
                    "word": " our",
                    "probability": 0.330078125
                },
                {
                    "start": 13.6,
                    "end": 14.0,
                    "word": " differential",
                    "probability": 0.7197265625
                },
                {
                    "start": 14.0,
                    "end": 14.5,
                    "word": " privacy",
                    "probability": 0.81103515625
                },
                {
                    "start": 14.5,
                    "end": 15.06,
                    "word": ".",
                    "probability": 0.984375
                }
            ]
        },
        {
            "id": 4,
            "seek": 3000,
            "start": 15.42,
            "end": 19.26,
            "text": " Before we start, let me play this video for you.",
            "tokens": [
                51114,
                4546,
                321,
                722,
                11,
                718,
                385,
                862,
                341,
                960,
                337,
                291,
                13,
                51314
            ],
            "temperature": 0.0,
            "avg_logprob": -0.28414948453608246,
            "compression_ratio": 1.6422413793103448,
            "no_speech_prob": 0.4638671875,
            "words": [
                {
                    "start": 15.42,
                    "end": 15.72,
                    "word": " Before",
                    "probability": 0.83056640625
                },
                {
                    "start": 15.72,
                    "end": 15.96,
                    "word": " we",
                    "probability": 0.900390625
                },
                {
                    "start": 15.96,
                    "end": 16.2,
                    "word": " start",
                    "probability": 0.86865234375
                },
                {
                    "start": 16.2,
                    "end": 16.56,
                    "word": ",",
                    "probability": 0.93505859375
                },
                {
                    "start": 16.56,
                    "end": 16.72,
                    "word": " let",
                    "probability": 0.8916015625
                },
                {
                    "start": 16.72,
                    "end": 17.02,
                    "word": " me",
                    "probability": 0.90673828125
                },
                {
                    "start": 17.02,
                    "end": 18.16,
                    "word": " play",
                    "probability": 0.8623046875
                },
                {
                    "start": 18.16,
                    "end": 18.38,
                    "word": " this",
                    "probability": 0.87890625
                },
                {
                    "start": 18.38,
                    "end": 18.56,
                    "word": " video",
                    "probability": 0.79541015625
                },
                {
                    "start": 18.56,
                    "end": 18.78,
                    "word": " for",
                    "probability": 0.89013671875
                },
                {
                    "start": 18.78,
                    "end": 18.98,
                    "word": " you",
                    "probability": 0.896484375
                },
                {
                    "start": 18.98,
                    "end": 19.26,
                    "word": ".",
                    "probability": 0.9912109375
                }
            ]
        },
        {
            "id": 5,
            "seek": 3000,
            "start": 21.14,
            "end": 25.26,
            "text": " At Google, we use data to make our services work better for you.",
            "tokens": [
                51414,
                1711,
                3329,
                11,
                321,
                764,
                1412,
                281,
                652,
                527,
                3328,
                589,
                1101,
                337,
                291,
                13,
                51564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.28414948453608246,
            "compression_ratio": 1.6422413793103448,
            "no_speech_prob": 0.4638671875,
            "words": [
                {
                    "start": 21.14,
                    "end": 21.52,
                    "word": " At",
                    "probability": 0.66259765625
                },
                {
                    "start": 21.52,
                    "end": 21.8,
                    "word": " Google",
                    "probability": 0.86865234375
                },
                {
                    "start": 21.8,
                    "end": 22.0,
                    "word": ",",
                    "probability": 0.85888671875
                },
                {
                    "start": 22.0,
                    "end": 22.18,
                    "word": " we",
                    "probability": 0.8994140625
                },
                {
                    "start": 22.18,
                    "end": 22.42,
                    "word": " use",
                    "probability": 0.8173828125
                },
                {
                    "start": 22.42,
                    "end": 22.74,
                    "word": " data",
                    "probability": 0.8232421875
                },
                {
                    "start": 22.74,
                    "end": 22.92,
                    "word": " to",
                    "probability": 0.90185546875
                },
                {
                    "start": 22.92,
                    "end": 23.22,
                    "word": " make",
                    "probability": 0.8681640625
                },
                {
                    "start": 23.22,
                    "end": 23.3,
                    "word": " our",
                    "probability": 0.80615234375
                },
                {
                    "start": 23.3,
                    "end": 23.88,
                    "word": " services",
                    "probability": 0.8916015625
                },
                {
                    "start": 23.88,
                    "end": 24.2,
                    "word": " work",
                    "probability": 0.84814453125
                },
                {
                    "start": 24.2,
                    "end": 24.46,
                    "word": " better",
                    "probability": 0.818359375
                },
                {
                    "start": 24.46,
                    "end": 24.66,
                    "word": " for",
                    "probability": 0.896484375
                },
                {
                    "start": 24.66,
                    "end": 24.88,
                    "word": " you",
                    "probability": 0.89404296875
                },
                {
                    "start": 24.88,
                    "end": 25.26,
                    "word": ".",
                    "probability": 0.7666015625
                }
            ]
        },
        {
            "id": 6,
            "seek": 3000,
            "start": 25.26,
            "end": 29.88,
            "text": " And it's our responsibility to keep your data private, safe, and secure.",
            "tokens": [
                51614,
                400,
                309,
                311,
                527,
                6357,
                281,
                1066,
                428,
                1412,
                4551,
                11,
                3273,
                11,
                293,
                7144,
                13,
                51814
            ],
            "temperature": 0.0,
            "avg_logprob": -0.28414948453608246,
            "compression_ratio": 1.6422413793103448,
            "no_speech_prob": 0.4638671875,
            "words": [
                {
                    "start": 25.26,
                    "end": 25.52,
                    "word": " And",
                    "probability": 0.78564453125
                },
                {
                    "start": 25.52,
                    "end": 25.72,
                    "word": " it's",
                    "probability": 0.943115234375
                },
                {
                    "start": 25.72,
                    "end": 26.02,
                    "word": " our",
                    "probability": 0.8154296875
                },
                {
                    "start": 26.02,
                    "end": 26.64,
                    "word": " responsibility",
                    "probability": 0.89453125
                },
                {
                    "start": 26.64,
                    "end": 27.16,
                    "word": " to",
                    "probability": 0.908203125
                },
                {
                    "start": 27.16,
                    "end": 27.38,
                    "word": " keep",
                    "probability": 0.8134765625
                },
                {
                    "start": 27.38,
                    "end": 27.54,
                    "word": " your",
                    "probability": 0.79443359375
                },
                {
                    "start": 27.54,
                    "end": 27.78,
                    "word": " data",
                    "probability": 0.8134765625
                },
                {
                    "start": 27.78,
                    "end": 28.28,
                    "word": " private",
                    "probability": 0.79248046875
                },
                {
                    "start": 28.28,
                    "end": 28.6,
                    "word": ",",
                    "probability": 0.966796875
                },
                {
                    "start": 28.6,
                    "end": 28.86,
                    "word": " safe",
                    "probability": 0.88427734375
                },
                {
                    "start": 28.86,
                    "end": 28.98,
                    "word": ",",
                    "probability": 0.55126953125
                },
                {
                    "start": 28.98,
                    "end": 29.2,
                    "word": " and",
                    "probability": 0.8994140625
                },
                {
                    "start": 29.2,
                    "end": 29.6,
                    "word": " secure",
                    "probability": 0.9140625
                },
                {
                    "start": 29.6,
                    "end": 29.88,
                    "word": ".",
                    "probability": 0.9853515625
                }
            ]
        },
        {
            "id": 7,
            "seek": 5734,
            "start": 30.0,
            "end": 34.88,
            "text": " That's why we continually invest in tools that protect individual privacy.",
            "tokens": [
                50364,
                663,
                311,
                983,
                321,
                22277,
                1963,
                294,
                3873,
                300,
                2371,
                2609,
                11427,
                13,
                50564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17236328140522042,
            "compression_ratio": 1.6590038314176245,
            "no_speech_prob": 0.00167083740234375,
            "words": [
                {
                    "start": 30.0,
                    "end": 30.66,
                    "word": " That's",
                    "probability": 0.86572265625
                },
                {
                    "start": 30.66,
                    "end": 30.78,
                    "word": " why",
                    "probability": 0.8134765625
                },
                {
                    "start": 30.78,
                    "end": 30.96,
                    "word": " we",
                    "probability": 0.90234375
                },
                {
                    "start": 30.96,
                    "end": 31.42,
                    "word": " continually",
                    "probability": 0.75341796875
                },
                {
                    "start": 31.42,
                    "end": 31.96,
                    "word": " invest",
                    "probability": 0.7861328125
                },
                {
                    "start": 31.96,
                    "end": 32.2,
                    "word": " in",
                    "probability": 0.908203125
                },
                {
                    "start": 32.2,
                    "end": 32.46,
                    "word": " tools",
                    "probability": 0.7783203125
                },
                {
                    "start": 32.46,
                    "end": 32.64,
                    "word": " that",
                    "probability": 0.87255859375
                },
                {
                    "start": 32.64,
                    "end": 33.02,
                    "word": " protect",
                    "probability": 0.77392578125
                },
                {
                    "start": 33.02,
                    "end": 33.62,
                    "word": " individual",
                    "probability": 0.73876953125
                },
                {
                    "start": 33.62,
                    "end": 34.2,
                    "word": " privacy",
                    "probability": 0.833984375
                },
                {
                    "start": 34.2,
                    "end": 34.88,
                    "word": ".",
                    "probability": 0.75146484375
                }
            ]
        },
        {
            "id": 8,
            "seek": 5734,
            "start": 35.02,
            "end": 38.32,
            "text": " Differentially private analysis is one of those tools.",
            "tokens": [
                50614,
                20825,
                2270,
                4551,
                5215,
                307,
                472,
                295,
                729,
                3873,
                13,
                50764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17236328140522042,
            "compression_ratio": 1.6590038314176245,
            "no_speech_prob": 0.00167083740234375,
            "words": [
                {
                    "start": 35.02,
                    "end": 35.64,
                    "word": " Differentially",
                    "probability": 0.6768798828125
                },
                {
                    "start": 35.64,
                    "end": 36.08,
                    "word": " private",
                    "probability": 0.489013671875
                },
                {
                    "start": 36.08,
                    "end": 36.68,
                    "word": " analysis",
                    "probability": 0.76123046875
                },
                {
                    "start": 36.68,
                    "end": 37.1,
                    "word": " is",
                    "probability": 0.9033203125
                },
                {
                    "start": 37.1,
                    "end": 37.28,
                    "word": " one",
                    "probability": 0.81884765625
                },
                {
                    "start": 37.28,
                    "end": 37.38,
                    "word": " of",
                    "probability": 0.9033203125
                },
                {
                    "start": 37.38,
                    "end": 37.56,
                    "word": " those",
                    "probability": 0.8359375
                },
                {
                    "start": 37.56,
                    "end": 37.98,
                    "word": " tools",
                    "probability": 0.7744140625
                },
                {
                    "start": 37.98,
                    "end": 38.32,
                    "word": ".",
                    "probability": 0.8994140625
                }
            ]
        },
        {
            "id": 9,
            "seek": 5734,
            "start": 39.1,
            "end": 46.98,
            "text": " Many of our features, like popular times and maps, are only possible because we can learn from the activity patterns of millions of users.",
            "tokens": [
                50814,
                5126,
                295,
                527,
                4122,
                11,
                411,
                3743,
                1413,
                293,
                11317,
                11,
                366,
                787,
                1944,
                570,
                321,
                393,
                1466,
                490,
                264,
                5191,
                8294,
                295,
                6803,
                295,
                5022,
                13,
                51164
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17236328140522042,
            "compression_ratio": 1.6590038314176245,
            "no_speech_prob": 0.00167083740234375,
            "words": [
                {
                    "start": 39.1,
                    "end": 39.38,
                    "word": " Many",
                    "probability": 0.88720703125
                },
                {
                    "start": 39.38,
                    "end": 39.52,
                    "word": " of",
                    "probability": 0.9052734375
                },
                {
                    "start": 39.52,
                    "end": 39.72,
                    "word": " our",
                    "probability": 0.8125
                },
                {
                    "start": 39.72,
                    "end": 40.14,
                    "word": " features",
                    "probability": 0.73583984375
                },
                {
                    "start": 40.14,
                    "end": 40.5,
                    "word": ",",
                    "probability": 0.468994140625
                },
                {
                    "start": 40.5,
                    "end": 40.72,
                    "word": " like",
                    "probability": 0.8681640625
                },
                {
                    "start": 40.72,
                    "end": 41.14,
                    "word": " popular",
                    "probability": 0.76025390625
                },
                {
                    "start": 41.14,
                    "end": 41.5,
                    "word": " times",
                    "probability": 0.75830078125
                },
                {
                    "start": 41.5,
                    "end": 41.68,
                    "word": " and",
                    "probability": 0.806640625
                },
                {
                    "start": 41.68,
                    "end": 42.04,
                    "word": " maps",
                    "probability": 0.74609375
                },
                {
                    "start": 42.04,
                    "end": 42.26,
                    "word": ",",
                    "probability": 0.8701171875
                },
                {
                    "start": 42.26,
                    "end": 42.38,
                    "word": " are",
                    "probability": 0.88427734375
                },
                {
                    "start": 42.38,
                    "end": 42.58,
                    "word": " only",
                    "probability": 0.85400390625
                },
                {
                    "start": 42.58,
                    "end": 43.04,
                    "word": " possible",
                    "probability": 0.84375
                },
                {
                    "start": 43.04,
                    "end": 43.4,
                    "word": " because",
                    "probability": 0.82470703125
                },
                {
                    "start": 43.4,
                    "end": 43.56,
                    "word": " we",
                    "probability": 0.90625
                },
                {
                    "start": 43.56,
                    "end": 43.7,
                    "word": " can",
                    "probability": 0.89794921875
                },
                {
                    "start": 43.7,
                    "end": 43.94,
                    "word": " learn",
                    "probability": 0.91162109375
                },
                {
                    "start": 43.94,
                    "end": 44.14,
                    "word": " from",
                    "probability": 0.8232421875
                },
                {
                    "start": 44.14,
                    "end": 44.28,
                    "word": " the",
                    "probability": 0.80712890625
                },
                {
                    "start": 44.28,
                    "end": 44.7,
                    "word": " activity",
                    "probability": 0.78271484375
                },
                {
                    "start": 44.7,
                    "end": 45.16,
                    "word": " patterns",
                    "probability": 0.751953125
                },
                {
                    "start": 45.16,
                    "end": 45.44,
                    "word": " of",
                    "probability": 0.8994140625
                },
                {
                    "start": 45.44,
                    "end": 45.8,
                    "word": " millions",
                    "probability": 0.84375
                },
                {
                    "start": 45.8,
                    "end": 46.06,
                    "word": " of",
                    "probability": 0.9091796875
                },
                {
                    "start": 46.06,
                    "end": 46.4,
                    "word": " users",
                    "probability": 0.84375
                },
                {
                    "start": 46.4,
                    "end": 46.98,
                    "word": ".",
                    "probability": 0.97216796875
                }
            ]
        },
        {
            "id": 10,
            "seek": 5734,
            "start": 46.98,
            "end": 54.68,
            "text": " With differential privacy, we can protect an individual's data while still preserving the usefulness of the data at scale.",
            "tokens": [
                51214,
                2022,
                15756,
                11427,
                11,
                321,
                393,
                2371,
                364,
                2609,
                311,
                1412,
                1339,
                920,
                33173,
                264,
                4420,
                1287,
                295,
                264,
                1412,
                412,
                4373,
                13,
                51564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17236328140522042,
            "compression_ratio": 1.6590038314176245,
            "no_speech_prob": 0.00167083740234375,
            "words": [
                {
                    "start": 46.98,
                    "end": 47.32,
                    "word": " With",
                    "probability": 0.890625
                },
                {
                    "start": 47.32,
                    "end": 47.8,
                    "word": " differential",
                    "probability": 0.62841796875
                },
                {
                    "start": 47.8,
                    "end": 48.24,
                    "word": " privacy",
                    "probability": 0.8349609375
                },
                {
                    "start": 48.24,
                    "end": 48.56,
                    "word": ",",
                    "probability": 0.74267578125
                },
                {
                    "start": 48.56,
                    "end": 48.88,
                    "word": " we",
                    "probability": 0.89990234375
                },
                {
                    "start": 48.88,
                    "end": 49.06,
                    "word": " can",
                    "probability": 0.892578125
                },
                {
                    "start": 49.06,
                    "end": 49.5,
                    "word": " protect",
                    "probability": 0.77490234375
                },
                {
                    "start": 49.5,
                    "end": 49.7,
                    "word": " an",
                    "probability": 0.8984375
                },
                {
                    "start": 49.7,
                    "end": 50.46,
                    "word": " individual's",
                    "probability": 0.8544921875
                },
                {
                    "start": 50.46,
                    "end": 50.72,
                    "word": " data",
                    "probability": 0.806640625
                },
                {
                    "start": 50.72,
                    "end": 51.18,
                    "word": " while",
                    "probability": 0.7216796875
                },
                {
                    "start": 51.18,
                    "end": 51.54,
                    "word": " still",
                    "probability": 0.89794921875
                },
                {
                    "start": 51.54,
                    "end": 52.08,
                    "word": " preserving",
                    "probability": 0.8359375
                },
                {
                    "start": 52.08,
                    "end": 52.4,
                    "word": " the",
                    "probability": 0.81982421875
                },
                {
                    "start": 52.4,
                    "end": 53.06,
                    "word": " usefulness",
                    "probability": 0.90869140625
                },
                {
                    "start": 53.06,
                    "end": 53.24,
                    "word": " of",
                    "probability": 0.90625
                },
                {
                    "start": 53.24,
                    "end": 53.36,
                    "word": " the",
                    "probability": 0.80859375
                },
                {
                    "start": 53.36,
                    "end": 53.62,
                    "word": " data",
                    "probability": 0.83154296875
                },
                {
                    "start": 53.62,
                    "end": 53.84,
                    "word": " at",
                    "probability": 0.8759765625
                },
                {
                    "start": 53.84,
                    "end": 54.2,
                    "word": " scale",
                    "probability": 0.78857421875
                },
                {
                    "start": 54.2,
                    "end": 54.68,
                    "word": ".",
                    "probability": 0.98193359375
                }
            ]
        },
        {
            "id": 11,
            "seek": 5734,
            "start": 54.68,
            "end": 57.34,
            "text": " Let's take a closer look at how it works.",
            "tokens": [
                51614,
                961,
                311,
                747,
                257,
                4966,
                574,
                412,
                577,
                309,
                1985,
                13,
                51714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17236328140522042,
            "compression_ratio": 1.6590038314176245,
            "no_speech_prob": 0.00167083740234375,
            "words": [
                {
                    "start": 54.68,
                    "end": 55.06,
                    "word": " Let's",
                    "probability": 0.947265625
                },
                {
                    "start": 55.06,
                    "end": 55.24,
                    "word": " take",
                    "probability": 0.8125
                },
                {
                    "start": 55.38,
                    "end": 55.62,
                    "word": " a closer",
                    "probability": 0.80859375
                },
                {
                    "start": 55.62,
                    "end": 55.98,
                    "word": " look",
                    "probability": 0.90283203125
                },
                {
                    "start": 55.98,
                    "end": 56.12,
                    "word": " at",
                    "probability": 0.9072265625
                },
                {
                    "start": 56.12,
                    "end": 56.28,
                    "word": " how",
                    "probability": 0.90234375
                },
                {
                    "start": 56.28,
                    "end": 56.42,
                    "word": " it",
                    "probability": 0.9052734375
                },
                {
                    "start": 56.42,
                    "end": 56.82,
                    "word": " works",
                    "probability": 0.7314453125
                },
                {
                    "start": 56.82,
                    "end": 57.34,
                    "word": ".",
                    "probability": 0.912109375
                }
            ]
        },
        {
            "id": 12,
            "seek": 7946,
            "start": 57.34,
            "end": 63.18,
            "text": " Imagine you're trying to plan a trip to the grocery store and want to know how busy it is at the moment.",
            "tokens": [
                50364,
                11739,
                291,
                434,
                1382,
                281,
                1393,
                257,
                4931,
                281,
                264,
                14410,
                3531,
                293,
                528,
                281,
                458,
                577,
                5856,
                309,
                307,
                412,
                264,
                1623,
                13,
                50614
            ],
            "temperature": 0.0,
            "avg_logprob": -0.16965662991559063,
            "compression_ratio": 1.6103286384976525,
            "no_speech_prob": 0.0018339157104492188,
            "words": [
                {
                    "start": 57.34,
                    "end": 57.72,
                    "word": " Imagine",
                    "probability": 0.70947265625
                },
                {
                    "start": 57.72,
                    "end": 57.98,
                    "word": " you're",
                    "probability": 0.85009765625
                },
                {
                    "start": 57.98,
                    "end": 58.24,
                    "word": " trying",
                    "probability": 0.736328125
                },
                {
                    "start": 58.24,
                    "end": 58.42,
                    "word": " to",
                    "probability": 0.908203125
                },
                {
                    "start": 58.42,
                    "end": 58.62,
                    "word": " plan",
                    "probability": 0.8779296875
                },
                {
                    "start": 58.78,
                    "end": 58.98,
                    "word": " a trip",
                    "probability": 0.80908203125
                },
                {
                    "start": 58.98,
                    "end": 59.14,
                    "word": " to",
                    "probability": 0.90234375
                },
                {
                    "start": 59.14,
                    "end": 59.28,
                    "word": " the",
                    "probability": 0.8017578125
                },
                {
                    "start": 59.28,
                    "end": 59.58,
                    "word": " grocery",
                    "probability": 0.65625
                },
                {
                    "start": 59.58,
                    "end": 60.06,
                    "word": " store",
                    "probability": 0.79833984375
                },
                {
                    "start": 60.06,
                    "end": 60.46,
                    "word": " and",
                    "probability": 0.76904296875
                },
                {
                    "start": 60.46,
                    "end": 60.68,
                    "word": " want",
                    "probability": 0.8017578125
                },
                {
                    "start": 60.68,
                    "end": 60.8,
                    "word": " to",
                    "probability": 0.90576171875
                },
                {
                    "start": 60.8,
                    "end": 60.98,
                    "word": " know",
                    "probability": 0.81396484375
                },
                {
                    "start": 60.98,
                    "end": 61.26,
                    "word": " how",
                    "probability": 0.9033203125
                },
                {
                    "start": 61.26,
                    "end": 61.64,
                    "word": " busy",
                    "probability": 0.86376953125
                },
                {
                    "start": 61.64,
                    "end": 61.9,
                    "word": " it",
                    "probability": 0.9052734375
                },
                {
                    "start": 61.9,
                    "end": 62.08,
                    "word": " is",
                    "probability": 0.904296875
                },
                {
                    "start": 62.08,
                    "end": 62.24,
                    "word": " at",
                    "probability": 0.9013671875
                },
                {
                    "start": 62.24,
                    "end": 62.38,
                    "word": " the",
                    "probability": 0.8046875
                },
                {
                    "start": 62.38,
                    "end": 62.66,
                    "word": " moment",
                    "probability": 0.912109375
                },
                {
                    "start": 62.66,
                    "end": 63.18,
                    "word": ".",
                    "probability": 0.88232421875
                }
            ]
        },
        {
            "id": 13,
            "seek": 7946,
            "start": 63.88,
            "end": 72.32,
            "text": " We use differentially private analysis on the historical and real-time data of other shoppers to show you how busy a place is.",
            "tokens": [
                50664,
                492,
                764,
                819,
                2270,
                4551,
                5215,
                322,
                264,
                8584,
                293,
                957,
                12,
                3766,
                1412,
                295,
                661,
                3945,
                21819,
                281,
                855,
                291,
                577,
                5856,
                257,
                1081,
                307,
                13,
                51064
            ],
            "temperature": 0.0,
            "avg_logprob": -0.16965662991559063,
            "compression_ratio": 1.6103286384976525,
            "no_speech_prob": 0.0018339157104492188,
            "words": [
                {
                    "start": 63.88,
                    "end": 63.96,
                    "word": " We",
                    "probability": 0.89794921875
                },
                {
                    "start": 63.96,
                    "end": 64.24,
                    "word": " use",
                    "probability": 0.7607421875
                },
                {
                    "start": 64.24,
                    "end": 64.9,
                    "word": " differentially",
                    "probability": 0.6064453125
                },
                {
                    "start": 64.9,
                    "end": 65.34,
                    "word": " private",
                    "probability": 0.79931640625
                },
                {
                    "start": 65.34,
                    "end": 65.92,
                    "word": " analysis",
                    "probability": 0.78466796875
                },
                {
                    "start": 65.92,
                    "end": 66.36,
                    "word": " on",
                    "probability": 0.8955078125
                },
                {
                    "start": 66.36,
                    "end": 66.48,
                    "word": " the",
                    "probability": 0.8056640625
                },
                {
                    "start": 66.48,
                    "end": 67.0,
                    "word": " historical",
                    "probability": 0.783203125
                },
                {
                    "start": 67.0,
                    "end": 67.62,
                    "word": " and",
                    "probability": 0.88427734375
                },
                {
                    "start": 67.62,
                    "end": 68.0,
                    "word": " real",
                    "probability": 0.86328125
                },
                {
                    "start": 68.0,
                    "end": 68.28,
                    "word": "-time",
                    "probability": 0.759033203125
                },
                {
                    "start": 68.28,
                    "end": 68.62,
                    "word": " data",
                    "probability": 0.810546875
                },
                {
                    "start": 68.62,
                    "end": 68.86,
                    "word": " of",
                    "probability": 0.90234375
                },
                {
                    "start": 68.86,
                    "end": 69.1,
                    "word": " other",
                    "probability": 0.79345703125
                },
                {
                    "start": 69.1,
                    "end": 69.7,
                    "word": " shoppers",
                    "probability": 0.85791015625
                },
                {
                    "start": 69.7,
                    "end": 70.06,
                    "word": " to",
                    "probability": 0.7958984375
                },
                {
                    "start": 70.06,
                    "end": 70.28,
                    "word": " show",
                    "probability": 0.88525390625
                },
                {
                    "start": 70.28,
                    "end": 70.48,
                    "word": " you",
                    "probability": 0.89208984375
                },
                {
                    "start": 70.48,
                    "end": 70.7,
                    "word": " how",
                    "probability": 0.89697265625
                },
                {
                    "start": 70.7,
                    "end": 71.04,
                    "word": " busy",
                    "probability": 0.85546875
                },
                {
                    "start": 71.22,
                    "end": 71.46,
                    "word": " a place",
                    "probability": 0.75732421875
                },
                {
                    "start": 71.46,
                    "end": 71.8,
                    "word": " is",
                    "probability": 0.90087890625
                },
                {
                    "start": 71.8,
                    "end": 72.32,
                    "word": ".",
                    "probability": 0.97021484375
                }
            ]
        },
        {
            "id": 14,
            "seek": 7946,
            "start": 72.72,
            "end": 79.46,
            "text": " The analysis adds noise to results so you can look at how crowded a place is without identifying who was there.",
            "tokens": [
                51114,
                440,
                5215,
                10860,
                5658,
                281,
                3542,
                370,
                291,
                393,
                574,
                412,
                577,
                21634,
                257,
                1081,
                307,
                1553,
                16696,
                567,
                390,
                456,
                13,
                51414
            ],
            "temperature": 0.0,
            "avg_logprob": -0.16965662991559063,
            "compression_ratio": 1.6103286384976525,
            "no_speech_prob": 0.0018339157104492188,
            "words": [
                {
                    "start": 72.72,
                    "end": 72.84,
                    "word": " The",
                    "probability": 0.77880859375
                },
                {
                    "start": 72.84,
                    "end": 73.28,
                    "word": " analysis",
                    "probability": 0.7529296875
                },
                {
                    "start": 73.28,
                    "end": 73.72,
                    "word": " adds",
                    "probability": 0.7158203125
                },
                {
                    "start": 73.72,
                    "end": 74.04,
                    "word": " noise",
                    "probability": 0.87646484375
                },
                {
                    "start": 74.04,
                    "end": 74.2,
                    "word": " to",
                    "probability": 0.8984375
                },
                {
                    "start": 74.2,
                    "end": 74.7,
                    "word": " results",
                    "probability": 0.68017578125
                },
                {
                    "start": 74.7,
                    "end": 75.04,
                    "word": " so",
                    "probability": 0.61279296875
                },
                {
                    "start": 75.04,
                    "end": 75.18,
                    "word": " you",
                    "probability": 0.89208984375
                },
                {
                    "start": 75.18,
                    "end": 75.34,
                    "word": " can",
                    "probability": 0.8974609375
                },
                {
                    "start": 75.34,
                    "end": 75.48,
                    "word": " look",
                    "probability": 0.9150390625
                },
                {
                    "start": 75.48,
                    "end": 75.64,
                    "word": " at",
                    "probability": 0.9052734375
                },
                {
                    "start": 75.64,
                    "end": 75.9,
                    "word": " how",
                    "probability": 0.89208984375
                },
                {
                    "start": 75.9,
                    "end": 76.32,
                    "word": " crowded",
                    "probability": 0.6484375
                },
                {
                    "start": 76.5,
                    "end": 76.74,
                    "word": " a place",
                    "probability": 0.767578125
                },
                {
                    "start": 76.74,
                    "end": 77.02,
                    "word": " is",
                    "probability": 0.90283203125
                },
                {
                    "start": 77.02,
                    "end": 77.46,
                    "word": " without",
                    "probability": 0.7939453125
                },
                {
                    "start": 77.46,
                    "end": 78.18,
                    "word": " identifying",
                    "probability": 0.91357421875
                },
                {
                    "start": 78.18,
                    "end": 78.42,
                    "word": " who",
                    "probability": 0.80078125
                },
                {
                    "start": 78.42,
                    "end": 78.58,
                    "word": " was",
                    "probability": 0.74267578125
                },
                {
                    "start": 78.58,
                    "end": 78.94,
                    "word": " there",
                    "probability": 0.8544921875
                },
                {
                    "start": 78.94,
                    "end": 79.46,
                    "word": ".",
                    "probability": 0.98095703125
                }
            ]
        },
        {
            "id": 15,
            "seek": 10424,
            "start": 79.46,
            "end": 91.24,
            "text": " Differentially private analysis is what enables us to show millions of our users every day how busy a place is while protecting people's privacy with confidence.",
            "tokens": [
                50364,
                20825,
                2270,
                4551,
                5215,
                307,
                437,
                17077,
                505,
                281,
                855,
                6803,
                295,
                527,
                5022,
                633,
                786,
                577,
                5856,
                257,
                1081,
                307,
                1339,
                12316,
                561,
                311,
                11427,
                365,
                6687,
                13,
                50914
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20301943030711767,
            "compression_ratio": 1.5485232067510548,
            "no_speech_prob": 0.09393310546875,
            "words": [
                {
                    "start": 79.46,
                    "end": 80.7,
                    "word": " Differentially",
                    "probability": 0.5004730224609375
                },
                {
                    "start": 80.7,
                    "end": 81.12,
                    "word": " private",
                    "probability": 0.380615234375
                },
                {
                    "start": 81.12,
                    "end": 81.72,
                    "word": " analysis",
                    "probability": 0.68994140625
                },
                {
                    "start": 81.72,
                    "end": 82.06,
                    "word": " is",
                    "probability": 0.87841796875
                },
                {
                    "start": 82.06,
                    "end": 82.38,
                    "word": " what",
                    "probability": 0.892578125
                },
                {
                    "start": 82.38,
                    "end": 82.82,
                    "word": " enables",
                    "probability": 0.76416015625
                },
                {
                    "start": 82.82,
                    "end": 83.14,
                    "word": " us",
                    "probability": 0.884765625
                },
                {
                    "start": 83.14,
                    "end": 83.36,
                    "word": " to",
                    "probability": 0.9072265625
                },
                {
                    "start": 83.36,
                    "end": 83.6,
                    "word": " show",
                    "probability": 0.8662109375
                },
                {
                    "start": 83.6,
                    "end": 84.1,
                    "word": " millions",
                    "probability": 0.845703125
                },
                {
                    "start": 84.1,
                    "end": 84.36,
                    "word": " of",
                    "probability": 0.9072265625
                },
                {
                    "start": 84.36,
                    "end": 84.62,
                    "word": " our",
                    "probability": 0.81494140625
                },
                {
                    "start": 84.62,
                    "end": 85.04,
                    "word": " users",
                    "probability": 0.8369140625
                },
                {
                    "start": 85.04,
                    "end": 85.56,
                    "word": " every",
                    "probability": 0.56640625
                },
                {
                    "start": 85.56,
                    "end": 86.04,
                    "word": " day",
                    "probability": 0.8994140625
                },
                {
                    "start": 86.04,
                    "end": 86.58,
                    "word": " how",
                    "probability": 0.81396484375
                },
                {
                    "start": 86.58,
                    "end": 86.98,
                    "word": " busy",
                    "probability": 0.84765625
                },
                {
                    "start": 87.14,
                    "end": 87.38,
                    "word": " a place",
                    "probability": 0.76806640625
                },
                {
                    "start": 87.38,
                    "end": 87.7,
                    "word": " is",
                    "probability": 0.89794921875
                },
                {
                    "start": 87.7,
                    "end": 88.2,
                    "word": " while",
                    "probability": 0.6474609375
                },
                {
                    "start": 88.2,
                    "end": 88.68,
                    "word": " protecting",
                    "probability": 0.7099609375
                },
                {
                    "start": 88.68,
                    "end": 89.18,
                    "word": " people's",
                    "probability": 0.9365234375
                },
                {
                    "start": 89.18,
                    "end": 89.56,
                    "word": " privacy",
                    "probability": 0.84814453125
                },
                {
                    "start": 89.56,
                    "end": 89.84,
                    "word": " with",
                    "probability": 0.8203125
                },
                {
                    "start": 89.84,
                    "end": 90.5,
                    "word": " confidence",
                    "probability": 0.90087890625
                },
                {
                    "start": 90.5,
                    "end": 91.24,
                    "word": ".",
                    "probability": 0.9453125
                }
            ]
        },
        {
            "id": 16,
            "seek": 10424,
            "start": 91.38,
            "end": 97.82,
            "text": " Many important research, business, and social questions can be answered by analyzing large data sets.",
            "tokens": [
                50964,
                5126,
                1021,
                2132,
                11,
                1606,
                11,
                293,
                2093,
                1651,
                393,
                312,
                10103,
                538,
                23663,
                2416,
                1412,
                6352,
                13,
                51214
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20301943030711767,
            "compression_ratio": 1.5485232067510548,
            "no_speech_prob": 0.09393310546875,
            "words": [
                {
                    "start": 91.38,
                    "end": 91.56,
                    "word": " Many",
                    "probability": 0.865234375
                },
                {
                    "start": 91.56,
                    "end": 91.98,
                    "word": " important",
                    "probability": 0.7333984375
                },
                {
                    "start": 91.98,
                    "end": 92.64,
                    "word": " research",
                    "probability": 0.798828125
                },
                {
                    "start": 92.64,
                    "end": 92.92,
                    "word": ",",
                    "probability": 0.8828125
                },
                {
                    "start": 92.92,
                    "end": 93.18,
                    "word": " business",
                    "probability": 0.88671875
                },
                {
                    "start": 93.18,
                    "end": 93.54,
                    "word": ",",
                    "probability": 0.60205078125
                },
                {
                    "start": 93.54,
                    "end": 93.78,
                    "word": " and",
                    "probability": 0.8994140625
                },
                {
                    "start": 93.78,
                    "end": 94.2,
                    "word": " social",
                    "probability": 0.79736328125
                },
                {
                    "start": 94.2,
                    "end": 94.68,
                    "word": " questions",
                    "probability": 0.91357421875
                },
                {
                    "start": 94.68,
                    "end": 94.86,
                    "word": " can",
                    "probability": 0.88037109375
                },
                {
                    "start": 94.86,
                    "end": 95.0,
                    "word": " be",
                    "probability": 0.9091796875
                },
                {
                    "start": 95.0,
                    "end": 95.36,
                    "word": " answered",
                    "probability": 0.7158203125
                },
                {
                    "start": 95.36,
                    "end": 95.64,
                    "word": " by",
                    "probability": 0.9228515625
                },
                {
                    "start": 95.64,
                    "end": 96.18,
                    "word": " analyzing",
                    "probability": 0.67919921875
                },
                {
                    "start": 96.18,
                    "end": 96.64,
                    "word": " large",
                    "probability": 0.927734375
                },
                {
                    "start": 96.64,
                    "end": 96.94,
                    "word": " data",
                    "probability": 0.495849609375
                },
                {
                    "start": 96.94,
                    "end": 97.42,
                    "word": " sets",
                    "probability": 0.7841796875
                },
                {
                    "start": 97.42,
                    "end": 97.82,
                    "word": ".",
                    "probability": 0.3330078125
                }
            ]
        },
        {
            "id": 17,
            "seek": 10424,
            "start": 97.82,
            "end": 104.24,
            "text": " Which is why, after years of development, Google has made its differential privacy library open source.",
            "tokens": [
                51264,
                3013,
                307,
                983,
                11,
                934,
                924,
                295,
                3250,
                11,
                3329,
                575,
                1027,
                1080,
                15756,
                11427,
                6405,
                1269,
                4009,
                13,
                51564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20301943030711767,
            "compression_ratio": 1.5485232067510548,
            "no_speech_prob": 0.09393310546875,
            "words": [
                {
                    "start": 97.82,
                    "end": 97.98,
                    "word": " Which",
                    "probability": 0.82861328125
                },
                {
                    "start": 97.98,
                    "end": 98.12,
                    "word": " is",
                    "probability": 0.90380859375
                },
                {
                    "start": 98.12,
                    "end": 98.34,
                    "word": " why",
                    "probability": 0.82080078125
                },
                {
                    "start": 98.34,
                    "end": 98.48,
                    "word": ",",
                    "probability": 0.47900390625
                },
                {
                    "start": 98.48,
                    "end": 98.64,
                    "word": " after",
                    "probability": 0.72265625
                },
                {
                    "start": 98.64,
                    "end": 99.0,
                    "word": " years",
                    "probability": 0.86572265625
                },
                {
                    "start": 99.0,
                    "end": 99.2,
                    "word": " of",
                    "probability": 0.90625
                },
                {
                    "start": 99.2,
                    "end": 99.64,
                    "word": " development",
                    "probability": 0.77197265625
                },
                {
                    "start": 99.64,
                    "end": 100.1,
                    "word": ",",
                    "probability": 0.9423828125
                },
                {
                    "start": 100.1,
                    "end": 100.28,
                    "word": " Google",
                    "probability": 0.85791015625
                },
                {
                    "start": 100.28,
                    "end": 100.58,
                    "word": " has",
                    "probability": 0.890625
                },
                {
                    "start": 100.58,
                    "end": 100.86,
                    "word": " made",
                    "probability": 0.81591796875
                },
                {
                    "start": 100.86,
                    "end": 101.1,
                    "word": " its",
                    "probability": 0.79248046875
                },
                {
                    "start": 101.1,
                    "end": 101.58,
                    "word": " differential",
                    "probability": 0.33837890625
                },
                {
                    "start": 101.58,
                    "end": 102.12,
                    "word": " privacy",
                    "probability": 0.84375
                },
                {
                    "start": 102.12,
                    "end": 102.72,
                    "word": " library",
                    "probability": 0.8154296875
                },
                {
                    "start": 102.72,
                    "end": 103.12,
                    "word": " open",
                    "probability": 0.72802734375
                },
                {
                    "start": 103.12,
                    "end": 103.5,
                    "word": " source",
                    "probability": 0.5771484375
                },
                {
                    "start": 103.5,
                    "end": 104.24,
                    "word": ".",
                    "probability": 0.97265625
                }
            ]
        },
        {
            "id": 18,
            "seek": 13152,
            "start": 104.24,
            "end": 118.66,
            "text": " Whether you are a researcher working on new medical treatments or a developer building an exciting app, this open source library can help you easily and affordably gain insights from data in a private, secure way.",
            "tokens": [
                50364,
                8503,
                291,
                366,
                257,
                21751,
                1364,
                322,
                777,
                4625,
                15795,
                420,
                257,
                10754,
                2390,
                364,
                4670,
                724,
                11,
                341,
                1269,
                4009,
                6405,
                393,
                854,
                291,
                3612,
                293,
                6157,
                1188,
                6052,
                14310,
                490,
                1412,
                294,
                257,
                4551,
                11,
                7144,
                636,
                13,
                51014
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18840842728697976,
            "compression_ratio": 1.608695652173913,
            "no_speech_prob": 0.09259033203125,
            "words": [
                {
                    "start": 104.24,
                    "end": 104.74,
                    "word": " Whether",
                    "probability": 0.6787109375
                },
                {
                    "start": 104.74,
                    "end": 104.96,
                    "word": " you",
                    "probability": 0.89697265625
                },
                {
                    "start": 104.96,
                    "end": 105.14,
                    "word": " are",
                    "probability": 0.85400390625
                },
                {
                    "start": 105.42,
                    "end": 105.74,
                    "word": " a researcher",
                    "probability": 0.76318359375
                },
                {
                    "start": 105.74,
                    "end": 106.34,
                    "word": " working",
                    "probability": 0.68115234375
                },
                {
                    "start": 106.34,
                    "end": 106.6,
                    "word": " on",
                    "probability": 0.90576171875
                },
                {
                    "start": 106.6,
                    "end": 106.88,
                    "word": " new",
                    "probability": 0.83544921875
                },
                {
                    "start": 106.88,
                    "end": 107.18,
                    "word": " medical",
                    "probability": 0.90966796875
                },
                {
                    "start": 107.18,
                    "end": 107.78,
                    "word": " treatments",
                    "probability": 0.76806640625
                },
                {
                    "start": 107.78,
                    "end": 108.28,
                    "word": " or",
                    "probability": 0.57177734375
                },
                {
                    "start": 108.46,
                    "end": 108.86,
                    "word": " a developer",
                    "probability": 0.626953125
                },
                {
                    "start": 108.86,
                    "end": 109.5,
                    "word": " building",
                    "probability": 0.75634765625
                },
                {
                    "start": 109.5,
                    "end": 109.76,
                    "word": " an",
                    "probability": 0.8974609375
                },
                {
                    "start": 109.76,
                    "end": 110.22,
                    "word": " exciting",
                    "probability": 0.857421875
                },
                {
                    "start": 110.22,
                    "end": 110.62,
                    "word": " app",
                    "probability": 0.90869140625
                },
                {
                    "start": 110.62,
                    "end": 111.16,
                    "word": ",",
                    "probability": 0.662109375
                },
                {
                    "start": 111.16,
                    "end": 111.4,
                    "word": " this",
                    "probability": 0.85986328125
                },
                {
                    "start": 111.4,
                    "end": 111.8,
                    "word": " open",
                    "probability": 0.7392578125
                },
                {
                    "start": 111.8,
                    "end": 112.06,
                    "word": " source",
                    "probability": 0.54345703125
                },
                {
                    "start": 112.06,
                    "end": 112.52,
                    "word": " library",
                    "probability": 0.81103515625
                },
                {
                    "start": 112.52,
                    "end": 112.88,
                    "word": " can",
                    "probability": 0.8984375
                },
                {
                    "start": 112.88,
                    "end": 113.1,
                    "word": " help",
                    "probability": 0.8056640625
                },
                {
                    "start": 113.1,
                    "end": 113.28,
                    "word": " you",
                    "probability": 0.8916015625
                },
                {
                    "start": 113.28,
                    "end": 113.82,
                    "word": " easily",
                    "probability": 0.8642578125
                },
                {
                    "start": 113.82,
                    "end": 114.2,
                    "word": " and",
                    "probability": 0.88818359375
                },
                {
                    "start": 114.2,
                    "end": 114.78,
                    "word": " affordably",
                    "probability": 0.942626953125
                },
                {
                    "start": 114.78,
                    "end": 115.24,
                    "word": " gain",
                    "probability": 0.87646484375
                },
                {
                    "start": 115.24,
                    "end": 115.76,
                    "word": " insights",
                    "probability": 0.80615234375
                },
                {
                    "start": 115.76,
                    "end": 115.96,
                    "word": " from",
                    "probability": 0.79638671875
                },
                {
                    "start": 115.96,
                    "end": 116.24,
                    "word": " data",
                    "probability": 0.80908203125
                },
                {
                    "start": 116.24,
                    "end": 116.54,
                    "word": " in",
                    "probability": 0.89697265625
                },
                {
                    "start": 116.68,
                    "end": 117.1,
                    "word": " a private",
                    "probability": 0.80615234375
                },
                {
                    "start": 117.1,
                    "end": 117.3,
                    "word": ",",
                    "probability": 0.56005859375
                },
                {
                    "start": 117.3,
                    "end": 117.52,
                    "word": " secure",
                    "probability": 0.8896484375
                },
                {
                    "start": 117.52,
                    "end": 117.94,
                    "word": " way",
                    "probability": 0.89892578125
                },
                {
                    "start": 117.94,
                    "end": 118.66,
                    "word": ".",
                    "probability": 0.99365234375
                }
            ]
        },
        {
            "id": 19,
            "seek": 13152,
            "start": 118.92,
            "end": 125.2,
            "text": " Differential privacy is just one of many ways we build privacy protections into our products.",
            "tokens": [
                51064,
                413,
                12612,
                2549,
                11427,
                307,
                445,
                472,
                295,
                867,
                2098,
                321,
                1322,
                11427,
                29031,
                666,
                527,
                3383,
                13,
                51364
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18840842728697976,
            "compression_ratio": 1.608695652173913,
            "no_speech_prob": 0.09259033203125,
            "words": [
                {
                    "start": 118.92,
                    "end": 119.36,
                    "word": " Differential",
                    "probability": 0.9026692708333334
                },
                {
                    "start": 119.36,
                    "end": 119.78,
                    "word": " privacy",
                    "probability": 0.79296875
                },
                {
                    "start": 119.78,
                    "end": 120.22,
                    "word": " is",
                    "probability": 0.9033203125
                },
                {
                    "start": 120.22,
                    "end": 120.5,
                    "word": " just",
                    "probability": 0.8896484375
                },
                {
                    "start": 120.5,
                    "end": 120.94,
                    "word": " one",
                    "probability": 0.82080078125
                },
                {
                    "start": 120.94,
                    "end": 121.22,
                    "word": " of",
                    "probability": 0.8955078125
                },
                {
                    "start": 121.22,
                    "end": 121.58,
                    "word": " many",
                    "probability": 0.76806640625
                },
                {
                    "start": 121.58,
                    "end": 122.04,
                    "word": " ways",
                    "probability": 0.80908203125
                },
                {
                    "start": 122.04,
                    "end": 122.32,
                    "word": " we",
                    "probability": 0.900390625
                },
                {
                    "start": 122.32,
                    "end": 122.62,
                    "word": " build",
                    "probability": 0.8232421875
                },
                {
                    "start": 122.62,
                    "end": 123.06,
                    "word": " privacy",
                    "probability": 0.837890625
                },
                {
                    "start": 123.06,
                    "end": 123.58,
                    "word": " protections",
                    "probability": 0.87353515625
                },
                {
                    "start": 123.58,
                    "end": 124.12,
                    "word": " into",
                    "probability": 0.72265625
                },
                {
                    "start": 124.12,
                    "end": 124.4,
                    "word": " our",
                    "probability": 0.810546875
                },
                {
                    "start": 124.4,
                    "end": 124.98,
                    "word": " products",
                    "probability": 0.74072265625
                },
                {
                    "start": 124.98,
                    "end": 125.2,
                    "word": ".",
                    "probability": 0.458740234375
                }
            ]
        },
        {
            "id": 20,
            "seek": 13152,
            "start": 125.2,
            "end": 131.52,
            "text": " And we will continue to innovate to new technologies that keep your data private, safe, and secure.",
            "tokens": [
                51414,
                400,
                321,
                486,
                2354,
                281,
                33444,
                281,
                777,
                7943,
                300,
                1066,
                428,
                1412,
                4551,
                11,
                3273,
                11,
                293,
                7144,
                13,
                51714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18840842728697976,
            "compression_ratio": 1.608695652173913,
            "no_speech_prob": 0.09259033203125,
            "words": [
                {
                    "start": 125.2,
                    "end": 125.42,
                    "word": " And",
                    "probability": 0.86767578125
                },
                {
                    "start": 125.42,
                    "end": 125.56,
                    "word": " we",
                    "probability": 0.890625
                },
                {
                    "start": 125.56,
                    "end": 125.72,
                    "word": " will",
                    "probability": 0.80908203125
                },
                {
                    "start": 125.72,
                    "end": 126.08,
                    "word": " continue",
                    "probability": 0.794921875
                },
                {
                    "start": 126.08,
                    "end": 126.34,
                    "word": " to",
                    "probability": 0.90673828125
                },
                {
                    "start": 126.34,
                    "end": 126.7,
                    "word": " innovate",
                    "probability": 0.81005859375
                },
                {
                    "start": 126.7,
                    "end": 126.94,
                    "word": " to",
                    "probability": 0.83251953125
                },
                {
                    "start": 126.94,
                    "end": 127.18,
                    "word": " new",
                    "probability": 0.81982421875
                },
                {
                    "start": 127.18,
                    "end": 127.8,
                    "word": " technologies",
                    "probability": 0.89599609375
                },
                {
                    "start": 127.8,
                    "end": 128.26,
                    "word": " that",
                    "probability": 0.8759765625
                },
                {
                    "start": 128.26,
                    "end": 128.52,
                    "word": " keep",
                    "probability": 0.82080078125
                },
                {
                    "start": 128.52,
                    "end": 128.68,
                    "word": " your",
                    "probability": 0.78466796875
                },
                {
                    "start": 128.68,
                    "end": 128.94,
                    "word": " data",
                    "probability": 0.80908203125
                },
                {
                    "start": 128.94,
                    "end": 129.48,
                    "word": " private",
                    "probability": 0.7783203125
                },
                {
                    "start": 129.48,
                    "end": 129.94,
                    "word": ",",
                    "probability": 0.9853515625
                },
                {
                    "start": 129.94,
                    "end": 130.26,
                    "word": " safe",
                    "probability": 0.8828125
                },
                {
                    "start": 130.26,
                    "end": 130.38,
                    "word": ",",
                    "probability": 0.55908203125
                },
                {
                    "start": 130.38,
                    "end": 130.72,
                    "word": " and",
                    "probability": 0.9013671875
                },
                {
                    "start": 130.72,
                    "end": 131.16,
                    "word": " secure",
                    "probability": 0.90087890625
                },
                {
                    "start": 131.16,
                    "end": 131.52,
                    "word": ".",
                    "probability": 0.99365234375
                }
            ]
        },
        {
            "id": 21,
            "seek": 16152,
            "start": 132.68,
            "end": 144.32,
            "text": " Great. Now that we've watched the video, I'll get into trying to talk a little bit more from a high level about differential privacy.",
            "tokens": [
                50414,
                3769,
                13,
                823,
                300,
                321,
                600,
                6337,
                264,
                960,
                11,
                286,
                603,
                483,
                666,
                1382,
                281,
                751,
                257,
                707,
                857,
                544,
                490,
                257,
                1090,
                1496,
                466,
                15756,
                11427,
                13,
                50964
            ],
            "temperature": 0.0,
            "avg_logprob": -0.23704769187851957,
            "compression_ratio": 1.485981308411215,
            "no_speech_prob": 0.032562255859375,
            "words": [
                {
                    "start": 132.68,
                    "end": 133.46,
                    "word": " Great",
                    "probability": 0.7080078125
                },
                {
                    "start": 133.46,
                    "end": 134.24,
                    "word": ".",
                    "probability": 0.6044921875
                },
                {
                    "start": 134.96,
                    "end": 135.34,
                    "word": " Now",
                    "probability": 0.85888671875
                },
                {
                    "start": 135.34,
                    "end": 135.52,
                    "word": " that",
                    "probability": 0.37744140625
                },
                {
                    "start": 135.52,
                    "end": 135.72,
                    "word": " we've",
                    "probability": 0.884033203125
                },
                {
                    "start": 135.72,
                    "end": 135.96,
                    "word": " watched",
                    "probability": 0.73193359375
                },
                {
                    "start": 135.96,
                    "end": 136.12,
                    "word": " the",
                    "probability": 0.81591796875
                },
                {
                    "start": 136.12,
                    "end": 136.48,
                    "word": " video",
                    "probability": 0.7958984375
                },
                {
                    "start": 136.48,
                    "end": 137.16,
                    "word": ",",
                    "probability": 0.853515625
                },
                {
                    "start": 137.16,
                    "end": 137.44,
                    "word": " I'll",
                    "probability": 0.902099609375
                },
                {
                    "start": 137.44,
                    "end": 137.68,
                    "word": " get",
                    "probability": 0.873046875
                },
                {
                    "start": 137.68,
                    "end": 138.34,
                    "word": " into",
                    "probability": 0.689453125
                },
                {
                    "start": 138.34,
                    "end": 138.76,
                    "word": " trying",
                    "probability": 0.77685546875
                },
                {
                    "start": 138.76,
                    "end": 139.0,
                    "word": " to",
                    "probability": 0.90283203125
                },
                {
                    "start": 139.0,
                    "end": 139.46,
                    "word": " talk",
                    "probability": 0.8115234375
                },
                {
                    "start": 139.76,
                    "end": 139.92,
                    "word": " a little",
                    "probability": 0.8037109375
                },
                {
                    "start": 139.92,
                    "end": 140.12,
                    "word": " bit",
                    "probability": 0.8505859375
                },
                {
                    "start": 140.12,
                    "end": 140.42,
                    "word": " more",
                    "probability": 0.88720703125
                },
                {
                    "start": 140.42,
                    "end": 141.0,
                    "word": " from",
                    "probability": 0.73779296875
                },
                {
                    "start": 141.14,
                    "end": 141.36,
                    "word": " a high",
                    "probability": 0.84521484375
                },
                {
                    "start": 141.36,
                    "end": 141.8,
                    "word": " level",
                    "probability": 0.7138671875
                },
                {
                    "start": 141.8,
                    "end": 142.62,
                    "word": " about",
                    "probability": 0.603515625
                },
                {
                    "start": 142.62,
                    "end": 143.04,
                    "word": " differential",
                    "probability": 0.70458984375
                },
                {
                    "start": 143.04,
                    "end": 143.54,
                    "word": " privacy",
                    "probability": 0.849609375
                },
                {
                    "start": 143.54,
                    "end": 144.32,
                    "word": ".",
                    "probability": 0.99267578125
                }
            ]
        },
        {
            "id": 22,
            "seek": 16152,
            "start": 144.86,
            "end": 152.76,
            "text": " The naive question that we always get is, why don't we just remove identifiers to anonymize data?",
            "tokens": [
                51014,
                440,
                29052,
                1168,
                300,
                321,
                1009,
                483,
                307,
                11,
                983,
                500,
                380,
                321,
                445,
                4159,
                2473,
                23463,
                281,
                37293,
                1125,
                1412,
                30,
                51364
            ],
            "temperature": 0.0,
            "avg_logprob": -0.23704769187851957,
            "compression_ratio": 1.485981308411215,
            "no_speech_prob": 0.032562255859375,
            "words": [
                {
                    "start": 144.86,
                    "end": 145.58,
                    "word": " The",
                    "probability": 0.77099609375
                },
                {
                    "start": 145.58,
                    "end": 145.86,
                    "word": " naive",
                    "probability": 0.323486328125
                },
                {
                    "start": 145.86,
                    "end": 146.22,
                    "word": " question",
                    "probability": 0.78564453125
                },
                {
                    "start": 146.22,
                    "end": 146.38,
                    "word": " that",
                    "probability": 0.63037109375
                },
                {
                    "start": 146.38,
                    "end": 146.56,
                    "word": " we",
                    "probability": 0.9033203125
                },
                {
                    "start": 146.56,
                    "end": 147.18,
                    "word": " always",
                    "probability": 0.76220703125
                },
                {
                    "start": 147.18,
                    "end": 147.46,
                    "word": " get",
                    "probability": 0.89208984375
                },
                {
                    "start": 147.46,
                    "end": 147.88,
                    "word": " is",
                    "probability": 0.89990234375
                },
                {
                    "start": 147.88,
                    "end": 148.38,
                    "word": ",",
                    "probability": 0.5771484375
                },
                {
                    "start": 148.38,
                    "end": 148.94,
                    "word": " why",
                    "probability": 0.599609375
                },
                {
                    "start": 148.94,
                    "end": 149.06,
                    "word": " don't",
                    "probability": 0.947021484375
                },
                {
                    "start": 149.06,
                    "end": 149.22,
                    "word": " we",
                    "probability": 0.90625
                },
                {
                    "start": 149.22,
                    "end": 149.4,
                    "word": " just",
                    "probability": 0.875
                },
                {
                    "start": 149.4,
                    "end": 149.8,
                    "word": " remove",
                    "probability": 0.857421875
                },
                {
                    "start": 149.8,
                    "end": 150.6,
                    "word": " identifiers",
                    "probability": 0.84912109375
                },
                {
                    "start": 150.6,
                    "end": 150.94,
                    "word": " to",
                    "probability": 0.88427734375
                },
                {
                    "start": 150.94,
                    "end": 151.58,
                    "word": " anonymize",
                    "probability": 0.841064453125
                },
                {
                    "start": 151.58,
                    "end": 151.98,
                    "word": " data",
                    "probability": 0.7626953125
                },
                {
                    "start": 151.98,
                    "end": 152.76,
                    "word": "?",
                    "probability": 0.91748046875
                }
            ]
        },
        {
            "id": 23,
            "seek": 16152,
            "start": 153.36,
            "end": 161.5,
            "text": " Unfortunately, removing identifiers might not be good enough, as well as removing PII.",
            "tokens": [
                51414,
                8590,
                11,
                12720,
                2473,
                23463,
                1062,
                406,
                312,
                665,
                1547,
                11,
                382,
                731,
                382,
                12720,
                430,
                9503,
                13,
                51814
            ],
            "temperature": 0.0,
            "avg_logprob": -0.23704769187851957,
            "compression_ratio": 1.485981308411215,
            "no_speech_prob": 0.032562255859375,
            "words": [
                {
                    "start": 153.36,
                    "end": 154.12,
                    "word": " Unfortunately",
                    "probability": 0.82421875
                },
                {
                    "start": 154.12,
                    "end": 154.36,
                    "word": ",",
                    "probability": 0.94580078125
                },
                {
                    "start": 154.36,
                    "end": 154.7,
                    "word": " removing",
                    "probability": 0.7939453125
                },
                {
                    "start": 154.7,
                    "end": 155.36,
                    "word": " identifiers",
                    "probability": 0.825439453125
                },
                {
                    "start": 155.36,
                    "end": 155.92,
                    "word": " might",
                    "probability": 0.77734375
                },
                {
                    "start": 155.92,
                    "end": 156.14,
                    "word": " not",
                    "probability": 0.89306640625
                },
                {
                    "start": 156.14,
                    "end": 156.34,
                    "word": " be",
                    "probability": 0.90673828125
                },
                {
                    "start": 156.34,
                    "end": 156.56,
                    "word": " good",
                    "probability": 0.88134765625
                },
                {
                    "start": 156.56,
                    "end": 156.94,
                    "word": " enough",
                    "probability": 0.794921875
                },
                {
                    "start": 156.94,
                    "end": 157.62,
                    "word": ",",
                    "probability": 0.49658203125
                },
                {
                    "start": 157.62,
                    "end": 158.92,
                    "word": " as",
                    "probability": 0.90966796875
                },
                {
                    "start": 158.92,
                    "end": 159.5,
                    "word": " well",
                    "probability": 0.8759765625
                },
                {
                    "start": 159.5,
                    "end": 160.16,
                    "word": " as",
                    "probability": 0.9140625
                },
                {
                    "start": 160.16,
                    "end": 160.56,
                    "word": " removing",
                    "probability": 0.7470703125
                },
                {
                    "start": 160.56,
                    "end": 161.38,
                    "word": " PII",
                    "probability": 0.93603515625
                },
                {
                    "start": 161.38,
                    "end": 161.5,
                    "word": ".",
                    "probability": 0.90869140625
                }
            ]
        },
        {
            "id": 24,
            "seek": 18668,
            "start": 161.52,
            "end": 168.6,
            "text": " There are multiple well-known data breaches resulting from publishing the quote-unquote de-identified data.",
            "tokens": [
                50364,
                821,
                366,
                3866,
                731,
                12,
                6861,
                1412,
                1403,
                13272,
                16505,
                490,
                17832,
                264,
                6513,
                12,
                409,
                25016,
                368,
                12,
                40613,
                1412,
                13,
                50664
            ],
            "temperature": 0.0,
            "avg_logprob": -0.22636718694120644,
            "compression_ratio": 1.641025641025641,
            "no_speech_prob": 0.025177001953125,
            "words": [
                {
                    "start": 161.52,
                    "end": 162.16,
                    "word": " There",
                    "probability": 0.275146484375
                },
                {
                    "start": 162.16,
                    "end": 162.3,
                    "word": " are",
                    "probability": 0.88818359375
                },
                {
                    "start": 162.3,
                    "end": 162.7,
                    "word": " multiple",
                    "probability": 0.736328125
                },
                {
                    "start": 162.7,
                    "end": 163.28,
                    "word": " well",
                    "probability": 0.83935546875
                },
                {
                    "start": 163.28,
                    "end": 163.5,
                    "word": "-known",
                    "probability": 0.723388671875
                },
                {
                    "start": 163.5,
                    "end": 163.74,
                    "word": " data",
                    "probability": 0.78515625
                },
                {
                    "start": 163.74,
                    "end": 164.04,
                    "word": " breaches",
                    "probability": 0.88623046875
                },
                {
                    "start": 164.04,
                    "end": 164.54,
                    "word": " resulting",
                    "probability": 0.83740234375
                },
                {
                    "start": 164.54,
                    "end": 164.9,
                    "word": " from",
                    "probability": 0.8134765625
                },
                {
                    "start": 164.9,
                    "end": 165.4,
                    "word": " publishing",
                    "probability": 0.732421875
                },
                {
                    "start": 165.4,
                    "end": 165.86,
                    "word": " the",
                    "probability": 0.328369140625
                },
                {
                    "start": 165.86,
                    "end": 166.38,
                    "word": " quote",
                    "probability": 0.30712890625
                },
                {
                    "start": 166.38,
                    "end": 166.8,
                    "word": "-unquote",
                    "probability": 0.7508138020833334
                },
                {
                    "start": 167.12,
                    "end": 167.4,
                    "word": " de-identified",
                    "probability": 0.863525390625
                },
                {
                    "start": 167.4,
                    "end": 167.94,
                    "word": " data",
                    "probability": 0.818359375
                },
                {
                    "start": 167.94,
                    "end": 168.6,
                    "word": ".",
                    "probability": 0.9619140625
                }
            ]
        },
        {
            "id": 25,
            "seek": 18668,
            "start": 168.6,
            "end": 175.24,
            "text": " One classical example involves the publication of de-identified medical records in the state of Washington.",
            "tokens": [
                50714,
                1485,
                13735,
                1365,
                11626,
                264,
                19953,
                295,
                368,
                12,
                40613,
                4625,
                7724,
                294,
                264,
                1785,
                295,
                6149,
                13,
                51014
            ],
            "temperature": 0.0,
            "avg_logprob": -0.22636718694120644,
            "compression_ratio": 1.641025641025641,
            "no_speech_prob": 0.025177001953125,
            "words": [
                {
                    "start": 168.6,
                    "end": 168.72,
                    "word": " One",
                    "probability": 0.53564453125
                },
                {
                    "start": 168.72,
                    "end": 169.08,
                    "word": " classical",
                    "probability": 0.74609375
                },
                {
                    "start": 169.08,
                    "end": 169.6,
                    "word": " example",
                    "probability": 0.94091796875
                },
                {
                    "start": 169.6,
                    "end": 170.1,
                    "word": " involves",
                    "probability": 0.76904296875
                },
                {
                    "start": 170.1,
                    "end": 170.36,
                    "word": " the",
                    "probability": 0.8076171875
                },
                {
                    "start": 170.36,
                    "end": 170.88,
                    "word": " publication",
                    "probability": 0.84130859375
                },
                {
                    "start": 170.88,
                    "end": 171.48,
                    "word": " of",
                    "probability": 0.904296875
                },
                {
                    "start": 171.7,
                    "end": 172.0,
                    "word": " de-identified",
                    "probability": 0.932373046875
                },
                {
                    "start": 172.0,
                    "end": 172.4,
                    "word": " medical",
                    "probability": 0.97509765625
                },
                {
                    "start": 172.4,
                    "end": 172.98,
                    "word": " records",
                    "probability": 0.78173828125
                },
                {
                    "start": 172.98,
                    "end": 173.72,
                    "word": " in",
                    "probability": 0.89111328125
                },
                {
                    "start": 173.72,
                    "end": 173.9,
                    "word": " the",
                    "probability": 0.8193359375
                },
                {
                    "start": 173.9,
                    "end": 174.04,
                    "word": " state",
                    "probability": 0.75146484375
                },
                {
                    "start": 174.04,
                    "end": 174.22,
                    "word": " of",
                    "probability": 0.91162109375
                },
                {
                    "start": 174.22,
                    "end": 174.66,
                    "word": " Washington",
                    "probability": 0.8369140625
                },
                {
                    "start": 174.66,
                    "end": 175.24,
                    "word": ".",
                    "probability": 0.94091796875
                }
            ]
        },
        {
            "id": 26,
            "seek": 18668,
            "start": 175.24,
            "end": 181.36,
            "text": " Those records could be partially joined with the voting registry using birthday and postcode.",
            "tokens": [
                51064,
                3950,
                7724,
                727,
                312,
                18886,
                6869,
                365,
                264,
                10419,
                36468,
                1228,
                6154,
                293,
                2183,
                22332,
                13,
                51314
            ],
            "temperature": 0.0,
            "avg_logprob": -0.22636718694120644,
            "compression_ratio": 1.641025641025641,
            "no_speech_prob": 0.025177001953125,
            "words": [
                {
                    "start": 175.24,
                    "end": 175.5,
                    "word": " Those",
                    "probability": 0.91064453125
                },
                {
                    "start": 175.5,
                    "end": 175.88,
                    "word": " records",
                    "probability": 0.775390625
                },
                {
                    "start": 175.88,
                    "end": 176.1,
                    "word": " could",
                    "probability": 0.806640625
                },
                {
                    "start": 176.1,
                    "end": 176.28,
                    "word": " be",
                    "probability": 0.91015625
                },
                {
                    "start": 176.28,
                    "end": 177.14,
                    "word": " partially",
                    "probability": 0.583984375
                },
                {
                    "start": 177.14,
                    "end": 177.7,
                    "word": " joined",
                    "probability": 0.87744140625
                },
                {
                    "start": 177.7,
                    "end": 178.06,
                    "word": " with",
                    "probability": 0.8056640625
                },
                {
                    "start": 178.06,
                    "end": 178.26,
                    "word": " the",
                    "probability": 0.63916015625
                },
                {
                    "start": 178.26,
                    "end": 178.46,
                    "word": " voting",
                    "probability": 0.67919921875
                },
                {
                    "start": 178.46,
                    "end": 178.98,
                    "word": " registry",
                    "probability": 0.6435546875
                },
                {
                    "start": 178.98,
                    "end": 179.46,
                    "word": " using",
                    "probability": 0.74072265625
                },
                {
                    "start": 179.46,
                    "end": 179.9,
                    "word": " birthday",
                    "probability": 0.316650390625
                },
                {
                    "start": 179.9,
                    "end": 180.12,
                    "word": " and",
                    "probability": 0.90380859375
                },
                {
                    "start": 180.12,
                    "end": 180.62,
                    "word": " postcode",
                    "probability": 0.761474609375
                },
                {
                    "start": 180.62,
                    "end": 181.36,
                    "word": ".",
                    "probability": 0.98486328125
                }
            ]
        },
        {
            "id": 27,
            "seek": 18668,
            "start": 181.36,
            "end": 186.68,
            "text": " This was sufficient to identify a majority of individuals in the data set.",
            "tokens": [
                51364,
                639,
                390,
                11563,
                281,
                5876,
                257,
                6286,
                295,
                5346,
                294,
                264,
                1412,
                992,
                13,
                51564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.22636718694120644,
            "compression_ratio": 1.641025641025641,
            "no_speech_prob": 0.025177001953125,
            "words": [
                {
                    "start": 181.36,
                    "end": 181.72,
                    "word": " This",
                    "probability": 0.8154296875
                },
                {
                    "start": 181.72,
                    "end": 182.06,
                    "word": " was",
                    "probability": 0.88671875
                },
                {
                    "start": 182.06,
                    "end": 182.44,
                    "word": " sufficient",
                    "probability": 0.8232421875
                },
                {
                    "start": 182.44,
                    "end": 182.66,
                    "word": " to",
                    "probability": 0.9072265625
                },
                {
                    "start": 182.66,
                    "end": 183.06,
                    "word": " identify",
                    "probability": 0.90625
                },
                {
                    "start": 183.34,
                    "end": 183.68,
                    "word": " a majority",
                    "probability": 0.83935546875
                },
                {
                    "start": 183.68,
                    "end": 184.04,
                    "word": " of",
                    "probability": 0.9033203125
                },
                {
                    "start": 184.04,
                    "end": 184.74,
                    "word": " individuals",
                    "probability": 0.6953125
                },
                {
                    "start": 184.74,
                    "end": 185.22,
                    "word": " in",
                    "probability": 0.90283203125
                },
                {
                    "start": 185.22,
                    "end": 185.38,
                    "word": " the",
                    "probability": 0.7978515625
                },
                {
                    "start": 185.38,
                    "end": 185.58,
                    "word": " data",
                    "probability": 0.49951171875
                },
                {
                    "start": 185.58,
                    "end": 185.92,
                    "word": " set",
                    "probability": 0.86865234375
                },
                {
                    "start": 185.92,
                    "end": 186.68,
                    "word": ".",
                    "probability": 0.99072265625
                }
            ]
        },
        {
            "id": 28,
            "seek": 21548,
            "start": 186.68,
            "end": 196.26,
            "text": " In another classical example, as a part of a competition in 2006 that intended to improve the movie recommendation system,",
            "tokens": [
                50364,
                682,
                1071,
                13735,
                1365,
                11,
                382,
                257,
                644,
                295,
                257,
                6211,
                294,
                14062,
                300,
                10226,
                281,
                3470,
                264,
                3169,
                11879,
                1185,
                11,
                50814
            ],
            "temperature": 0.0,
            "avg_logprob": -0.22395833333333334,
            "compression_ratio": 1.63671875,
            "no_speech_prob": 0.004608154296875,
            "words": [
                {
                    "start": 186.68,
                    "end": 187.04,
                    "word": " In",
                    "probability": 0.77392578125
                },
                {
                    "start": 187.04,
                    "end": 187.3,
                    "word": " another",
                    "probability": 0.8017578125
                },
                {
                    "start": 187.3,
                    "end": 187.82,
                    "word": " classical",
                    "probability": 0.79638671875
                },
                {
                    "start": 187.82,
                    "end": 188.42,
                    "word": " example",
                    "probability": 0.92431640625
                },
                {
                    "start": 188.42,
                    "end": 188.96,
                    "word": ",",
                    "probability": 0.79052734375
                },
                {
                    "start": 188.96,
                    "end": 189.12,
                    "word": " as",
                    "probability": 0.9033203125
                },
                {
                    "start": 189.26,
                    "end": 189.42,
                    "word": " a part",
                    "probability": 0.82080078125
                },
                {
                    "start": 189.42,
                    "end": 189.6,
                    "word": " of",
                    "probability": 0.904296875
                },
                {
                    "start": 190.1,
                    "end": 190.48,
                    "word": " a competition",
                    "probability": 0.939453125
                },
                {
                    "start": 190.48,
                    "end": 190.72,
                    "word": " in",
                    "probability": 0.87158203125
                },
                {
                    "start": 190.72,
                    "end": 191.82,
                    "word": " 2006",
                    "probability": 0.8603515625
                },
                {
                    "start": 191.82,
                    "end": 192.64,
                    "word": " that",
                    "probability": 0.681640625
                },
                {
                    "start": 192.64,
                    "end": 193.0,
                    "word": " intended",
                    "probability": 0.75537109375
                },
                {
                    "start": 193.0,
                    "end": 193.28,
                    "word": " to",
                    "probability": 0.90234375
                },
                {
                    "start": 193.28,
                    "end": 193.76,
                    "word": " improve",
                    "probability": 0.7646484375
                },
                {
                    "start": 193.76,
                    "end": 194.18,
                    "word": " the",
                    "probability": 0.7978515625
                },
                {
                    "start": 194.18,
                    "end": 194.4,
                    "word": " movie",
                    "probability": 0.74658203125
                },
                {
                    "start": 194.4,
                    "end": 195.02,
                    "word": " recommendation",
                    "probability": 0.47119140625
                },
                {
                    "start": 195.02,
                    "end": 195.72,
                    "word": " system",
                    "probability": 0.87060546875
                },
                {
                    "start": 195.72,
                    "end": 196.26,
                    "word": ",",
                    "probability": 0.86865234375
                }
            ]
        },
        {
            "id": 29,
            "seek": 21548,
            "start": 196.26,
            "end": 202.24,
            "text": " Netflix published movie ratings from its users. The records had the user identifiers removed,",
            "tokens": [
                50864,
                12778,
                6572,
                3169,
                24603,
                490,
                1080,
                5022,
                13,
                440,
                7724,
                632,
                264,
                4195,
                2473,
                23463,
                7261,
                11,
                51114
            ],
            "temperature": 0.0,
            "avg_logprob": -0.22395833333333334,
            "compression_ratio": 1.63671875,
            "no_speech_prob": 0.004608154296875,
            "words": [
                {
                    "start": 196.26,
                    "end": 196.52,
                    "word": " Netflix",
                    "probability": 0.78466796875
                },
                {
                    "start": 196.52,
                    "end": 197.1,
                    "word": " published",
                    "probability": 0.755859375
                },
                {
                    "start": 197.1,
                    "end": 197.52,
                    "word": " movie",
                    "probability": 0.806640625
                },
                {
                    "start": 197.52,
                    "end": 197.94,
                    "word": " ratings",
                    "probability": 0.828125
                },
                {
                    "start": 197.94,
                    "end": 198.24,
                    "word": " from",
                    "probability": 0.81103515625
                },
                {
                    "start": 198.24,
                    "end": 198.46,
                    "word": " its",
                    "probability": 0.8046875
                },
                {
                    "start": 198.46,
                    "end": 198.86,
                    "word": " users",
                    "probability": 0.86083984375
                },
                {
                    "start": 198.86,
                    "end": 199.54,
                    "word": ".",
                    "probability": 0.9150390625
                },
                {
                    "start": 199.62,
                    "end": 199.7,
                    "word": " The",
                    "probability": 0.7822265625
                },
                {
                    "start": 199.7,
                    "end": 199.96,
                    "word": " records",
                    "probability": 0.755859375
                },
                {
                    "start": 199.96,
                    "end": 200.26,
                    "word": " had",
                    "probability": 0.51953125
                },
                {
                    "start": 200.26,
                    "end": 200.42,
                    "word": " the",
                    "probability": 0.7578125
                },
                {
                    "start": 200.42,
                    "end": 200.68,
                    "word": " user",
                    "probability": 0.76806640625
                },
                {
                    "start": 200.68,
                    "end": 201.2,
                    "word": " identifiers",
                    "probability": 0.7744140625
                },
                {
                    "start": 201.2,
                    "end": 201.7,
                    "word": " removed",
                    "probability": 0.81591796875
                },
                {
                    "start": 201.7,
                    "end": 202.24,
                    "word": ",",
                    "probability": 0.79931640625
                }
            ]
        },
        {
            "id": 30,
            "seek": 21548,
            "start": 202.24,
            "end": 209.18,
            "text": " but researchers could partially join these records with IMDB databases, which were tied to identity.",
            "tokens": [
                51164,
                457,
                10309,
                727,
                18886,
                3917,
                613,
                7724,
                365,
                21463,
                27735,
                22380,
                11,
                597,
                645,
                9601,
                281,
                6575,
                13,
                51464
            ],
            "temperature": 0.0,
            "avg_logprob": -0.22395833333333334,
            "compression_ratio": 1.63671875,
            "no_speech_prob": 0.004608154296875,
            "words": [
                {
                    "start": 202.24,
                    "end": 202.4,
                    "word": " but",
                    "probability": 0.884765625
                },
                {
                    "start": 202.4,
                    "end": 202.88,
                    "word": " researchers",
                    "probability": 0.7509765625
                },
                {
                    "start": 202.88,
                    "end": 203.18,
                    "word": " could",
                    "probability": 0.80126953125
                },
                {
                    "start": 203.18,
                    "end": 203.52,
                    "word": " partially",
                    "probability": 0.77099609375
                },
                {
                    "start": 203.52,
                    "end": 203.9,
                    "word": " join",
                    "probability": 0.77197265625
                },
                {
                    "start": 203.9,
                    "end": 204.08,
                    "word": " these",
                    "probability": 0.708984375
                },
                {
                    "start": 204.08,
                    "end": 204.44,
                    "word": " records",
                    "probability": 0.77197265625
                },
                {
                    "start": 204.44,
                    "end": 204.68,
                    "word": " with",
                    "probability": 0.814453125
                },
                {
                    "start": 204.68,
                    "end": 205.78,
                    "word": " IMDB",
                    "probability": 0.789306640625
                },
                {
                    "start": 205.78,
                    "end": 206.5,
                    "word": " databases",
                    "probability": 0.91943359375
                },
                {
                    "start": 206.5,
                    "end": 207.14,
                    "word": ",",
                    "probability": 0.416015625
                },
                {
                    "start": 207.14,
                    "end": 207.36,
                    "word": " which",
                    "probability": 0.91259765625
                },
                {
                    "start": 207.36,
                    "end": 207.54,
                    "word": " were",
                    "probability": 0.87451171875
                },
                {
                    "start": 207.54,
                    "end": 207.86,
                    "word": " tied",
                    "probability": 0.8173828125
                },
                {
                    "start": 207.86,
                    "end": 208.04,
                    "word": " to",
                    "probability": 0.9091796875
                },
                {
                    "start": 208.04,
                    "end": 208.5,
                    "word": " identity",
                    "probability": 0.798828125
                },
                {
                    "start": 208.5,
                    "end": 209.18,
                    "word": ".",
                    "probability": 0.9609375
                }
            ]
        },
        {
            "id": 31,
            "seek": 21548,
            "start": 209.54,
            "end": 215.48,
            "text": " Such a join made possible to partially tie the full history of movie ranking to the user identifiers.",
            "tokens": [
                51514,
                9653,
                257,
                3917,
                1027,
                1944,
                281,
                18886,
                7582,
                264,
                1577,
                2503,
                295,
                3169,
                17833,
                281,
                264,
                4195,
                2473,
                23463,
                13,
                51764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.22395833333333334,
            "compression_ratio": 1.63671875,
            "no_speech_prob": 0.004608154296875,
            "words": [
                {
                    "start": 209.54,
                    "end": 209.8,
                    "word": " Such",
                    "probability": 0.822265625
                },
                {
                    "start": 209.98,
                    "end": 210.14,
                    "word": " a join",
                    "probability": 0.52978515625
                },
                {
                    "start": 210.14,
                    "end": 210.32,
                    "word": " made",
                    "probability": 0.7705078125
                },
                {
                    "start": 210.32,
                    "end": 210.74,
                    "word": " possible",
                    "probability": 0.56787109375
                },
                {
                    "start": 210.74,
                    "end": 211.02,
                    "word": " to",
                    "probability": 0.89599609375
                },
                {
                    "start": 211.02,
                    "end": 211.46,
                    "word": " partially",
                    "probability": 0.79736328125
                },
                {
                    "start": 211.46,
                    "end": 211.82,
                    "word": " tie",
                    "probability": 0.84765625
                },
                {
                    "start": 211.82,
                    "end": 211.98,
                    "word": " the",
                    "probability": 0.81396484375
                },
                {
                    "start": 211.98,
                    "end": 212.2,
                    "word": " full",
                    "probability": 0.85009765625
                },
                {
                    "start": 212.2,
                    "end": 212.48,
                    "word": " history",
                    "probability": 0.83544921875
                },
                {
                    "start": 212.48,
                    "end": 212.7,
                    "word": " of",
                    "probability": 0.90478515625
                },
                {
                    "start": 212.7,
                    "end": 212.98,
                    "word": " movie",
                    "probability": 0.77197265625
                },
                {
                    "start": 212.98,
                    "end": 213.4,
                    "word": " ranking",
                    "probability": 0.56298828125
                },
                {
                    "start": 213.4,
                    "end": 213.74,
                    "word": " to",
                    "probability": 0.8896484375
                },
                {
                    "start": 213.74,
                    "end": 213.88,
                    "word": " the",
                    "probability": 0.7587890625
                },
                {
                    "start": 213.88,
                    "end": 214.14,
                    "word": " user",
                    "probability": 0.78564453125
                },
                {
                    "start": 214.14,
                    "end": 214.8,
                    "word": " identifiers",
                    "probability": 0.805908203125
                },
                {
                    "start": 214.8,
                    "end": 215.48,
                    "word": ".",
                    "probability": 0.98779296875
                }
            ]
        },
        {
            "id": 32,
            "seek": 24548,
            "start": 215.48,
            "end": 222.54,
            "text": " This is a classical example of joining de-identified data sets and re-identifying some of the users.",
            "tokens": [
                50364,
                639,
                307,
                257,
                13735,
                1365,
                295,
                5549,
                368,
                12,
                40613,
                1412,
                6352,
                293,
                319,
                12,
                1078,
                5489,
                512,
                295,
                264,
                5022,
                13,
                50664
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17331861623603365,
            "compression_ratio": 1.804255319148936,
            "no_speech_prob": 0.00482940673828125,
            "words": [
                {
                    "start": 215.48,
                    "end": 215.96,
                    "word": " This",
                    "probability": 0.689453125
                },
                {
                    "start": 215.96,
                    "end": 216.12,
                    "word": " is",
                    "probability": 0.8125
                },
                {
                    "start": 216.18,
                    "end": 216.52,
                    "word": " a classical",
                    "probability": 0.81982421875
                },
                {
                    "start": 216.52,
                    "end": 217.06,
                    "word": " example",
                    "probability": 0.92578125
                },
                {
                    "start": 217.06,
                    "end": 217.82,
                    "word": " of",
                    "probability": 0.8955078125
                },
                {
                    "start": 217.82,
                    "end": 218.1,
                    "word": " joining",
                    "probability": 0.88720703125
                },
                {
                    "start": 218.38,
                    "end": 218.72,
                    "word": " de-identified",
                    "probability": 0.787353515625
                },
                {
                    "start": 218.72,
                    "end": 219.12,
                    "word": " data",
                    "probability": 0.263671875
                },
                {
                    "start": 219.12,
                    "end": 219.56,
                    "word": " sets",
                    "probability": 0.78955078125
                },
                {
                    "start": 219.56,
                    "end": 220.16,
                    "word": " and",
                    "probability": 0.845703125
                },
                {
                    "start": 220.16,
                    "end": 220.34,
                    "word": " re",
                    "probability": 0.91162109375
                },
                {
                    "start": 220.34,
                    "end": 220.88,
                    "word": "-identifying",
                    "probability": 0.9093424479166666
                },
                {
                    "start": 220.88,
                    "end": 221.24,
                    "word": " some",
                    "probability": 0.8017578125
                },
                {
                    "start": 221.24,
                    "end": 221.4,
                    "word": " of",
                    "probability": 0.87353515625
                },
                {
                    "start": 221.4,
                    "end": 221.54,
                    "word": " the",
                    "probability": 0.80029296875
                },
                {
                    "start": 221.54,
                    "end": 221.86,
                    "word": " users",
                    "probability": 0.83544921875
                },
                {
                    "start": 221.86,
                    "end": 222.54,
                    "word": ".",
                    "probability": 0.96435546875
                }
            ]
        },
        {
            "id": 33,
            "seek": 24548,
            "start": 222.68,
            "end": 229.56,
            "text": " Finally, de-identified search history can also, in certain cases, reveal the identity of a user.",
            "tokens": [
                50714,
                6288,
                11,
                368,
                12,
                40613,
                3164,
                2503,
                393,
                611,
                11,
                294,
                1629,
                3331,
                11,
                10658,
                264,
                6575,
                295,
                257,
                4195,
                13,
                51014
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17331861623603365,
            "compression_ratio": 1.804255319148936,
            "no_speech_prob": 0.00482940673828125,
            "words": [
                {
                    "start": 222.68,
                    "end": 223.18,
                    "word": " Finally",
                    "probability": 0.70947265625
                },
                {
                    "start": 223.18,
                    "end": 223.6,
                    "word": ",",
                    "probability": 0.91796875
                },
                {
                    "start": 223.62,
                    "end": 223.94,
                    "word": " de-identified",
                    "probability": 0.908447265625
                },
                {
                    "start": 223.94,
                    "end": 224.64,
                    "word": " search",
                    "probability": 0.8623046875
                },
                {
                    "start": 224.64,
                    "end": 224.98,
                    "word": " history",
                    "probability": 0.82373046875
                },
                {
                    "start": 224.98,
                    "end": 225.26,
                    "word": " can",
                    "probability": 0.88916015625
                },
                {
                    "start": 225.26,
                    "end": 225.56,
                    "word": " also",
                    "probability": 0.80908203125
                },
                {
                    "start": 225.56,
                    "end": 225.74,
                    "word": ",",
                    "probability": 0.63134765625
                },
                {
                    "start": 225.74,
                    "end": 225.76,
                    "word": " in",
                    "probability": 0.9033203125
                },
                {
                    "start": 225.76,
                    "end": 226.0,
                    "word": " certain",
                    "probability": 0.77490234375
                },
                {
                    "start": 226.0,
                    "end": 226.42,
                    "word": " cases",
                    "probability": 0.80517578125
                },
                {
                    "start": 226.42,
                    "end": 226.84,
                    "word": ",",
                    "probability": 0.98046875
                },
                {
                    "start": 226.84,
                    "end": 227.18,
                    "word": " reveal",
                    "probability": 0.87060546875
                },
                {
                    "start": 227.18,
                    "end": 227.52,
                    "word": " the",
                    "probability": 0.79443359375
                },
                {
                    "start": 227.52,
                    "end": 227.9,
                    "word": " identity",
                    "probability": 0.8466796875
                },
                {
                    "start": 227.9,
                    "end": 228.3,
                    "word": " of",
                    "probability": 0.9033203125
                },
                {
                    "start": 228.56,
                    "end": 228.88,
                    "word": " a user",
                    "probability": 0.8193359375
                },
                {
                    "start": 228.88,
                    "end": 229.56,
                    "word": ".",
                    "probability": 0.982421875
                }
            ]
        },
        {
            "id": 34,
            "seek": 24548,
            "start": 229.66,
            "end": 236.6,
            "text": " So, there is a pattern here. Even if the records are published without direct identifying information,",
            "tokens": [
                51064,
                407,
                11,
                456,
                307,
                257,
                5102,
                510,
                13,
                2754,
                498,
                264,
                7724,
                366,
                6572,
                1553,
                2047,
                16696,
                1589,
                11,
                51364
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17331861623603365,
            "compression_ratio": 1.804255319148936,
            "no_speech_prob": 0.00482940673828125,
            "words": [
                {
                    "start": 229.66,
                    "end": 229.94,
                    "word": " So",
                    "probability": 0.77099609375
                },
                {
                    "start": 229.94,
                    "end": 230.2,
                    "word": ",",
                    "probability": 0.497314453125
                },
                {
                    "start": 230.2,
                    "end": 230.58,
                    "word": " there",
                    "probability": 0.85009765625
                },
                {
                    "start": 230.58,
                    "end": 230.7,
                    "word": " is",
                    "probability": 0.69482421875
                },
                {
                    "start": 230.84,
                    "end": 231.18,
                    "word": " a pattern",
                    "probability": 0.767578125
                },
                {
                    "start": 231.18,
                    "end": 231.58,
                    "word": " here",
                    "probability": 0.80615234375
                },
                {
                    "start": 231.58,
                    "end": 232.14,
                    "word": ".",
                    "probability": 0.869140625
                },
                {
                    "start": 232.14,
                    "end": 232.4,
                    "word": " Even",
                    "probability": 0.8125
                },
                {
                    "start": 232.4,
                    "end": 232.58,
                    "word": " if",
                    "probability": 0.89892578125
                },
                {
                    "start": 232.58,
                    "end": 232.72,
                    "word": " the",
                    "probability": 0.79736328125
                },
                {
                    "start": 232.72,
                    "end": 232.98,
                    "word": " records",
                    "probability": 0.78076171875
                },
                {
                    "start": 232.98,
                    "end": 233.54,
                    "word": " are",
                    "probability": 0.8876953125
                },
                {
                    "start": 233.54,
                    "end": 233.98,
                    "word": " published",
                    "probability": 0.736328125
                },
                {
                    "start": 233.98,
                    "end": 234.36,
                    "word": " without",
                    "probability": 0.81396484375
                },
                {
                    "start": 234.36,
                    "end": 234.82,
                    "word": " direct",
                    "probability": 0.73876953125
                },
                {
                    "start": 234.82,
                    "end": 235.38,
                    "word": " identifying",
                    "probability": 0.94140625
                },
                {
                    "start": 235.38,
                    "end": 236.08,
                    "word": " information",
                    "probability": 0.7177734375
                },
                {
                    "start": 236.08,
                    "end": 236.6,
                    "word": ",",
                    "probability": 0.95654296875
                }
            ]
        },
        {
            "id": 35,
            "seek": 24548,
            "start": 236.6,
            "end": 245.46,
            "text": " the content of those records together with some auxiliary information may make re-identification possible in certain cases.",
            "tokens": [
                51414,
                264,
                2701,
                295,
                729,
                7724,
                1214,
                365,
                512,
                43741,
                1589,
                815,
                652,
                319,
                12,
                1078,
                3774,
                1944,
                294,
                1629,
                3331,
                13,
                51814
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17331861623603365,
            "compression_ratio": 1.804255319148936,
            "no_speech_prob": 0.00482940673828125,
            "words": [
                {
                    "start": 236.6,
                    "end": 236.76,
                    "word": " the",
                    "probability": 0.80859375
                },
                {
                    "start": 236.76,
                    "end": 237.04,
                    "word": " content",
                    "probability": 0.8134765625
                },
                {
                    "start": 237.04,
                    "end": 237.24,
                    "word": " of",
                    "probability": 0.90771484375
                },
                {
                    "start": 237.24,
                    "end": 237.44,
                    "word": " those",
                    "probability": 0.8369140625
                },
                {
                    "start": 237.44,
                    "end": 237.78,
                    "word": " records",
                    "probability": 0.79248046875
                },
                {
                    "start": 237.78,
                    "end": 238.3,
                    "word": " together",
                    "probability": 0.64013671875
                },
                {
                    "start": 238.3,
                    "end": 238.58,
                    "word": " with",
                    "probability": 0.8154296875
                },
                {
                    "start": 238.58,
                    "end": 238.8,
                    "word": " some",
                    "probability": 0.81689453125
                },
                {
                    "start": 238.8,
                    "end": 239.18,
                    "word": " auxiliary",
                    "probability": 0.857421875
                },
                {
                    "start": 239.18,
                    "end": 239.94,
                    "word": " information",
                    "probability": 0.72314453125
                },
                {
                    "start": 239.94,
                    "end": 240.6,
                    "word": " may",
                    "probability": 0.8271484375
                },
                {
                    "start": 240.6,
                    "end": 240.92,
                    "word": " make",
                    "probability": 0.58154296875
                },
                {
                    "start": 240.92,
                    "end": 242.8,
                    "word": " re",
                    "probability": 0.93505859375
                },
                {
                    "start": 242.8,
                    "end": 243.32,
                    "word": "-identification",
                    "probability": 0.9305013020833334
                },
                {
                    "start": 243.32,
                    "end": 243.84,
                    "word": " possible",
                    "probability": 0.861328125
                },
                {
                    "start": 243.84,
                    "end": 244.54,
                    "word": " in",
                    "probability": 0.841796875
                },
                {
                    "start": 244.54,
                    "end": 244.88,
                    "word": " certain",
                    "probability": 0.79150390625
                },
                {
                    "start": 244.88,
                    "end": 245.46,
                    "word": " cases",
                    "probability": 0.80126953125
                },
                {
                    "start": 245.46,
                    "end": 245.46,
                    "word": ".",
                    "probability": 0.98095703125
                }
            ]
        },
        {
            "id": 36,
            "seek": 27324,
            "start": 245.48,
            "end": 249.84,
            "text": " Thus, we need something better than de-identification in general.",
            "tokens": [
                50364,
                13827,
                11,
                321,
                643,
                746,
                1101,
                813,
                368,
                12,
                1078,
                3774,
                294,
                2674,
                13,
                50564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1771977953203432,
            "compression_ratio": 1.5450819672131149,
            "no_speech_prob": 0.0014781951904296875,
            "words": [
                {
                    "start": 245.48,
                    "end": 246.34,
                    "word": " Thus",
                    "probability": 0.413818359375
                },
                {
                    "start": 246.34,
                    "end": 246.42,
                    "word": ",",
                    "probability": 0.56005859375
                },
                {
                    "start": 246.42,
                    "end": 246.56,
                    "word": " we",
                    "probability": 0.88671875
                },
                {
                    "start": 246.56,
                    "end": 246.74,
                    "word": " need",
                    "probability": 0.75634765625
                },
                {
                    "start": 246.74,
                    "end": 247.02,
                    "word": " something",
                    "probability": 0.7255859375
                },
                {
                    "start": 247.02,
                    "end": 247.4,
                    "word": " better",
                    "probability": 0.8115234375
                },
                {
                    "start": 247.4,
                    "end": 247.64,
                    "word": " than",
                    "probability": 0.8720703125
                },
                {
                    "start": 247.84,
                    "end": 248.4,
                    "word": " de-identification",
                    "probability": 0.8946940104166666
                },
                {
                    "start": 248.4,
                    "end": 248.76,
                    "word": " in",
                    "probability": 0.85009765625
                },
                {
                    "start": 248.76,
                    "end": 249.24,
                    "word": " general",
                    "probability": 0.80224609375
                },
                {
                    "start": 249.24,
                    "end": 249.84,
                    "word": ".",
                    "probability": 0.9599609375
                }
            ]
        },
        {
            "id": 37,
            "seek": 27324,
            "start": 250.34,
            "end": 256.3,
            "text": " The second question that is very common to have is, well, why don't we just aggregate data?",
            "tokens": [
                50614,
                440,
                1150,
                1168,
                300,
                307,
                588,
                2689,
                281,
                362,
                307,
                11,
                731,
                11,
                983,
                500,
                380,
                321,
                445,
                26118,
                1412,
                30,
                50864
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1771977953203432,
            "compression_ratio": 1.5450819672131149,
            "no_speech_prob": 0.0014781951904296875,
            "words": [
                {
                    "start": 250.34,
                    "end": 250.52,
                    "word": " The",
                    "probability": 0.76171875
                },
                {
                    "start": 250.52,
                    "end": 250.72,
                    "word": " second",
                    "probability": 0.759765625
                },
                {
                    "start": 250.72,
                    "end": 251.14,
                    "word": " question",
                    "probability": 0.79248046875
                },
                {
                    "start": 251.14,
                    "end": 251.38,
                    "word": " that",
                    "probability": 0.8232421875
                },
                {
                    "start": 251.38,
                    "end": 252.06,
                    "word": " is",
                    "probability": 0.26171875
                },
                {
                    "start": 252.06,
                    "end": 252.28,
                    "word": " very",
                    "probability": 0.72216796875
                },
                {
                    "start": 252.28,
                    "end": 252.64,
                    "word": " common",
                    "probability": 0.7314453125
                },
                {
                    "start": 252.64,
                    "end": 252.82,
                    "word": " to",
                    "probability": 0.84619140625
                },
                {
                    "start": 252.82,
                    "end": 253.18,
                    "word": " have",
                    "probability": 0.87841796875
                },
                {
                    "start": 253.18,
                    "end": 253.54,
                    "word": " is",
                    "probability": 0.87939453125
                },
                {
                    "start": 253.54,
                    "end": 253.74,
                    "word": ",",
                    "probability": 0.62939453125
                },
                {
                    "start": 253.74,
                    "end": 253.86,
                    "word": " well",
                    "probability": 0.425048828125
                },
                {
                    "start": 253.86,
                    "end": 253.96,
                    "word": ",",
                    "probability": 0.92041015625
                },
                {
                    "start": 253.96,
                    "end": 254.2,
                    "word": " why",
                    "probability": 0.80322265625
                },
                {
                    "start": 254.2,
                    "end": 254.32,
                    "word": " don't",
                    "probability": 0.9365234375
                },
                {
                    "start": 254.32,
                    "end": 254.46,
                    "word": " we",
                    "probability": 0.90625
                },
                {
                    "start": 254.46,
                    "end": 254.68,
                    "word": " just",
                    "probability": 0.89599609375
                },
                {
                    "start": 254.68,
                    "end": 255.2,
                    "word": " aggregate",
                    "probability": 0.671875
                },
                {
                    "start": 255.2,
                    "end": 255.7,
                    "word": " data",
                    "probability": 0.7822265625
                },
                {
                    "start": 255.7,
                    "end": 256.3,
                    "word": "?",
                    "probability": 0.89013671875
                }
            ]
        },
        {
            "id": 38,
            "seek": 27324,
            "start": 256.56,
            "end": 264.68,
            "text": " In fact, it is a very well-known technique to publish aggregated statistics with buckets of users.",
            "tokens": [
                50914,
                682,
                1186,
                11,
                309,
                307,
                257,
                588,
                731,
                12,
                6861,
                6532,
                281,
                11374,
                16743,
                770,
                12523,
                365,
                32191,
                295,
                5022,
                13,
                51264
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1771977953203432,
            "compression_ratio": 1.5450819672131149,
            "no_speech_prob": 0.0014781951904296875,
            "words": [
                {
                    "start": 256.56,
                    "end": 256.84,
                    "word": " In",
                    "probability": 0.91552734375
                },
                {
                    "start": 256.84,
                    "end": 257.28,
                    "word": " fact",
                    "probability": 0.810546875
                },
                {
                    "start": 257.28,
                    "end": 257.56,
                    "word": ",",
                    "probability": 0.97998046875
                },
                {
                    "start": 257.56,
                    "end": 257.72,
                    "word": " it",
                    "probability": 0.88818359375
                },
                {
                    "start": 257.72,
                    "end": 257.84,
                    "word": " is",
                    "probability": 0.89013671875
                },
                {
                    "start": 257.98,
                    "end": 258.16,
                    "word": " a very",
                    "probability": 0.74462890625
                },
                {
                    "start": 258.16,
                    "end": 258.52,
                    "word": " well",
                    "probability": 0.86376953125
                },
                {
                    "start": 258.52,
                    "end": 258.86,
                    "word": "-known",
                    "probability": 0.732666015625
                },
                {
                    "start": 258.86,
                    "end": 259.58,
                    "word": " technique",
                    "probability": 0.9296875
                },
                {
                    "start": 259.58,
                    "end": 259.78,
                    "word": " to",
                    "probability": 0.88720703125
                },
                {
                    "start": 259.78,
                    "end": 260.1,
                    "word": " publish",
                    "probability": 0.86865234375
                },
                {
                    "start": 260.1,
                    "end": 260.72,
                    "word": " aggregated",
                    "probability": 0.755859375
                },
                {
                    "start": 260.72,
                    "end": 261.84,
                    "word": " statistics",
                    "probability": 0.8994140625
                },
                {
                    "start": 261.84,
                    "end": 262.74,
                    "word": " with",
                    "probability": 0.798828125
                },
                {
                    "start": 262.74,
                    "end": 263.54,
                    "word": " buckets",
                    "probability": 0.76318359375
                },
                {
                    "start": 263.54,
                    "end": 263.72,
                    "word": " of",
                    "probability": 0.90625
                },
                {
                    "start": 263.72,
                    "end": 264.08,
                    "word": " users",
                    "probability": 0.8544921875
                },
                {
                    "start": 264.08,
                    "end": 264.68,
                    "word": ".",
                    "probability": 0.984375
                }
            ]
        },
        {
            "id": 39,
            "seek": 27324,
            "start": 264.74,
            "end": 273.24,
            "text": " Usually, this involves grouping by a set of features. For example, grouping by restaurant visits, by location, and hour.",
            "tokens": [
                51314,
                11419,
                11,
                341,
                11626,
                40149,
                538,
                257,
                992,
                295,
                4122,
                13,
                1171,
                1365,
                11,
                40149,
                538,
                6383,
                17753,
                11,
                538,
                4914,
                11,
                293,
                1773,
                13,
                51714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1771977953203432,
            "compression_ratio": 1.5450819672131149,
            "no_speech_prob": 0.0014781951904296875,
            "words": [
                {
                    "start": 264.74,
                    "end": 265.26,
                    "word": " Usually",
                    "probability": 0.7763671875
                },
                {
                    "start": 265.26,
                    "end": 265.5,
                    "word": ",",
                    "probability": 0.77099609375
                },
                {
                    "start": 265.5,
                    "end": 265.64,
                    "word": " this",
                    "probability": 0.853515625
                },
                {
                    "start": 265.64,
                    "end": 265.96,
                    "word": " involves",
                    "probability": 0.841796875
                },
                {
                    "start": 265.96,
                    "end": 266.32,
                    "word": " grouping",
                    "probability": 0.81201171875
                },
                {
                    "start": 266.32,
                    "end": 266.58,
                    "word": " by",
                    "probability": 0.8974609375
                },
                {
                    "start": 266.76,
                    "end": 266.88,
                    "word": " a set",
                    "probability": 0.8955078125
                },
                {
                    "start": 266.88,
                    "end": 267.02,
                    "word": " of",
                    "probability": 0.9091796875
                },
                {
                    "start": 267.02,
                    "end": 267.4,
                    "word": " features",
                    "probability": 0.72314453125
                },
                {
                    "start": 267.4,
                    "end": 267.92,
                    "word": ".",
                    "probability": 0.6181640625
                },
                {
                    "start": 267.92,
                    "end": 268.1,
                    "word": " For",
                    "probability": 0.90478515625
                },
                {
                    "start": 268.1,
                    "end": 268.5,
                    "word": " example",
                    "probability": 0.93798828125
                },
                {
                    "start": 268.5,
                    "end": 269.0,
                    "word": ",",
                    "probability": 0.99560546875
                },
                {
                    "start": 269.0,
                    "end": 269.28,
                    "word": " grouping",
                    "probability": 0.8271484375
                },
                {
                    "start": 269.28,
                    "end": 269.76,
                    "word": " by",
                    "probability": 0.90234375
                },
                {
                    "start": 269.76,
                    "end": 270.68,
                    "word": " restaurant",
                    "probability": 0.77099609375
                },
                {
                    "start": 270.68,
                    "end": 271.16,
                    "word": " visits",
                    "probability": 0.88916015625
                },
                {
                    "start": 271.16,
                    "end": 271.28,
                    "word": ",",
                    "probability": 0.9501953125
                },
                {
                    "start": 271.28,
                    "end": 271.4,
                    "word": " by",
                    "probability": 0.890625
                },
                {
                    "start": 271.4,
                    "end": 271.9,
                    "word": " location",
                    "probability": 0.82470703125
                },
                {
                    "start": 271.9,
                    "end": 272.1,
                    "word": ",",
                    "probability": 0.55859375
                },
                {
                    "start": 272.1,
                    "end": 272.32,
                    "word": " and",
                    "probability": 0.8828125
                },
                {
                    "start": 272.32,
                    "end": 272.64,
                    "word": " hour",
                    "probability": 0.767578125
                },
                {
                    "start": 272.64,
                    "end": 273.24,
                    "word": ".",
                    "probability": 0.9892578125
                }
            ]
        },
        {
            "id": 40,
            "seek": 30054,
            "start": 273.24,
            "end": 282.26,
            "text": " This is commonly referred to as k-anonymity, since each user in the anonymized dataset is indistinguishable from at least k other users.",
            "tokens": [
                50364,
                639,
                307,
                12719,
                10839,
                281,
                382,
                350,
                12,
                282,
                12732,
                507,
                11,
                1670,
                1184,
                4195,
                294,
                264,
                37293,
                1602,
                28872,
                307,
                1016,
                468,
                7050,
                742,
                712,
                490,
                412,
                1935,
                350,
                661,
                5022,
                13,
                50764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20586340206185566,
            "compression_ratio": 1.5669291338582678,
            "no_speech_prob": 0.0007915496826171875,
            "words": [
                {
                    "start": 273.24,
                    "end": 273.64,
                    "word": " This",
                    "probability": 0.71484375
                },
                {
                    "start": 273.64,
                    "end": 273.78,
                    "word": " is",
                    "probability": 0.90966796875
                },
                {
                    "start": 273.78,
                    "end": 274.1,
                    "word": " commonly",
                    "probability": 0.724609375
                },
                {
                    "start": 274.1,
                    "end": 274.52,
                    "word": " referred",
                    "probability": 0.84619140625
                },
                {
                    "start": 274.52,
                    "end": 274.7,
                    "word": " to",
                    "probability": 0.9033203125
                },
                {
                    "start": 274.7,
                    "end": 275.0,
                    "word": " as",
                    "probability": 0.896484375
                },
                {
                    "start": 275.0,
                    "end": 275.24,
                    "word": " k",
                    "probability": 0.333251953125
                },
                {
                    "start": 275.24,
                    "end": 275.88,
                    "word": "-anonymity",
                    "probability": 0.8662109375
                },
                {
                    "start": 275.88,
                    "end": 276.26,
                    "word": ",",
                    "probability": 0.462158203125
                },
                {
                    "start": 276.26,
                    "end": 276.48,
                    "word": " since",
                    "probability": 0.8095703125
                },
                {
                    "start": 276.48,
                    "end": 276.78,
                    "word": " each",
                    "probability": 0.8671875
                },
                {
                    "start": 276.78,
                    "end": 277.1,
                    "word": " user",
                    "probability": 0.83544921875
                },
                {
                    "start": 277.1,
                    "end": 277.56,
                    "word": " in",
                    "probability": 0.88232421875
                },
                {
                    "start": 277.56,
                    "end": 277.94,
                    "word": " the",
                    "probability": 0.68896484375
                },
                {
                    "start": 277.94,
                    "end": 278.5,
                    "word": " anonymized",
                    "probability": 0.85205078125
                },
                {
                    "start": 278.5,
                    "end": 278.84,
                    "word": " dataset",
                    "probability": 0.6552734375
                },
                {
                    "start": 278.84,
                    "end": 279.16,
                    "word": " is",
                    "probability": 0.89111328125
                },
                {
                    "start": 279.16,
                    "end": 279.96,
                    "word": " indistinguishable",
                    "probability": 0.9685546875
                },
                {
                    "start": 279.96,
                    "end": 280.24,
                    "word": " from",
                    "probability": 0.80615234375
                },
                {
                    "start": 280.24,
                    "end": 280.44,
                    "word": " at",
                    "probability": 0.869140625
                },
                {
                    "start": 280.44,
                    "end": 280.6,
                    "word": " least",
                    "probability": 0.85595703125
                },
                {
                    "start": 280.6,
                    "end": 280.86,
                    "word": " k",
                    "probability": 0.92724609375
                },
                {
                    "start": 280.86,
                    "end": 281.14,
                    "word": " other",
                    "probability": 0.71142578125
                },
                {
                    "start": 281.14,
                    "end": 281.66,
                    "word": " users",
                    "probability": 0.85693359375
                },
                {
                    "start": 281.66,
                    "end": 282.26,
                    "word": ".",
                    "probability": 0.951171875
                }
            ]
        },
        {
            "id": 41,
            "seek": 30054,
            "start": 285.58,
            "end": 292.06,
            "text": " Now, let me explain to you why we need differential privacy and how it works.",
            "tokens": [
                50814,
                823,
                11,
                718,
                385,
                2903,
                281,
                291,
                983,
                321,
                643,
                15756,
                11427,
                293,
                577,
                309,
                1985,
                13,
                51164
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20586340206185566,
            "compression_ratio": 1.5669291338582678,
            "no_speech_prob": 0.0007915496826171875,
            "words": [
                {
                    "start": 285.58,
                    "end": 286.18,
                    "word": " Now",
                    "probability": 0.79296875
                },
                {
                    "start": 286.18,
                    "end": 286.78,
                    "word": ",",
                    "probability": 0.62548828125
                },
                {
                    "start": 286.78,
                    "end": 287.58,
                    "word": " let",
                    "probability": 0.876953125
                },
                {
                    "start": 287.58,
                    "end": 287.7,
                    "word": " me",
                    "probability": 0.90576171875
                },
                {
                    "start": 287.7,
                    "end": 288.08,
                    "word": " explain",
                    "probability": 0.82861328125
                },
                {
                    "start": 288.08,
                    "end": 289.06,
                    "word": " to",
                    "probability": 0.81298828125
                },
                {
                    "start": 289.06,
                    "end": 289.26,
                    "word": " you",
                    "probability": 0.8984375
                },
                {
                    "start": 289.26,
                    "end": 289.56,
                    "word": " why",
                    "probability": 0.7685546875
                },
                {
                    "start": 289.56,
                    "end": 289.76,
                    "word": " we",
                    "probability": 0.654296875
                },
                {
                    "start": 289.76,
                    "end": 289.9,
                    "word": " need",
                    "probability": 0.7265625
                },
                {
                    "start": 289.9,
                    "end": 290.24,
                    "word": " differential",
                    "probability": 0.72265625
                },
                {
                    "start": 290.24,
                    "end": 290.64,
                    "word": " privacy",
                    "probability": 0.8447265625
                },
                {
                    "start": 290.64,
                    "end": 290.86,
                    "word": " and",
                    "probability": 0.8623046875
                },
                {
                    "start": 290.86,
                    "end": 291.1,
                    "word": " how",
                    "probability": 0.89208984375
                },
                {
                    "start": 291.1,
                    "end": 291.24,
                    "word": " it",
                    "probability": 0.904296875
                },
                {
                    "start": 291.24,
                    "end": 291.52,
                    "word": " works",
                    "probability": 0.70947265625
                },
                {
                    "start": 291.52,
                    "end": 292.06,
                    "word": ".",
                    "probability": 0.97705078125
                }
            ]
        },
        {
            "id": 42,
            "seek": 30054,
            "start": 292.06,
            "end": 298.24,
            "text": " Let's imagine you are attending a two-day conference and the moderator shares some insights about the audience.",
            "tokens": [
                51214,
                961,
                311,
                3811,
                291,
                366,
                15862,
                257,
                732,
                12,
                810,
                7586,
                293,
                264,
                37778,
                12182,
                512,
                14310,
                466,
                264,
                4034,
                13,
                51464
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20586340206185566,
            "compression_ratio": 1.5669291338582678,
            "no_speech_prob": 0.0007915496826171875,
            "words": [
                {
                    "start": 292.06,
                    "end": 292.36,
                    "word": " Let's",
                    "probability": 0.93994140625
                },
                {
                    "start": 292.36,
                    "end": 292.7,
                    "word": " imagine",
                    "probability": 0.80859375
                },
                {
                    "start": 292.7,
                    "end": 293.0,
                    "word": " you",
                    "probability": 0.87109375
                },
                {
                    "start": 293.0,
                    "end": 293.12,
                    "word": " are",
                    "probability": 0.5009765625
                },
                {
                    "start": 293.12,
                    "end": 293.46,
                    "word": " attending",
                    "probability": 0.7822265625
                },
                {
                    "start": 293.76,
                    "end": 293.94,
                    "word": " a two",
                    "probability": 0.7265625
                },
                {
                    "start": 293.94,
                    "end": 294.06,
                    "word": "-day",
                    "probability": 0.836181640625
                },
                {
                    "start": 294.06,
                    "end": 294.48,
                    "word": " conference",
                    "probability": 0.72314453125
                },
                {
                    "start": 294.48,
                    "end": 294.82,
                    "word": " and",
                    "probability": 0.72509765625
                },
                {
                    "start": 294.82,
                    "end": 295.0,
                    "word": " the",
                    "probability": 0.7236328125
                },
                {
                    "start": 295.0,
                    "end": 295.82,
                    "word": " moderator",
                    "probability": 0.86865234375
                },
                {
                    "start": 295.82,
                    "end": 296.42,
                    "word": " shares",
                    "probability": 0.591796875
                },
                {
                    "start": 296.42,
                    "end": 296.74,
                    "word": " some",
                    "probability": 0.81103515625
                },
                {
                    "start": 296.74,
                    "end": 297.18,
                    "word": " insights",
                    "probability": 0.8193359375
                },
                {
                    "start": 297.18,
                    "end": 297.38,
                    "word": " about",
                    "probability": 0.59912109375
                },
                {
                    "start": 297.38,
                    "end": 297.54,
                    "word": " the",
                    "probability": 0.80908203125
                },
                {
                    "start": 297.54,
                    "end": 297.88,
                    "word": " audience",
                    "probability": 0.85986328125
                },
                {
                    "start": 297.88,
                    "end": 298.24,
                    "word": ".",
                    "probability": 0.98046875
                }
            ]
        },
        {
            "id": 43,
            "seek": 30054,
            "start": 298.34,
            "end": 302.58,
            "text": " Here, you can see the country of origin of the participants on day one.",
            "tokens": [
                51514,
                1692,
                11,
                291,
                393,
                536,
                264,
                1941,
                295,
                4957,
                295,
                264,
                10503,
                322,
                786,
                472,
                13,
                51664
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20586340206185566,
            "compression_ratio": 1.5669291338582678,
            "no_speech_prob": 0.0007915496826171875,
            "words": [
                {
                    "start": 298.34,
                    "end": 298.94,
                    "word": " Here",
                    "probability": 0.82666015625
                },
                {
                    "start": 298.94,
                    "end": 299.04,
                    "word": ",",
                    "probability": 0.4951171875
                },
                {
                    "start": 299.04,
                    "end": 299.16,
                    "word": " you",
                    "probability": 0.89404296875
                },
                {
                    "start": 299.16,
                    "end": 299.3,
                    "word": " can",
                    "probability": 0.89599609375
                },
                {
                    "start": 299.3,
                    "end": 299.44,
                    "word": " see",
                    "probability": 0.8095703125
                },
                {
                    "start": 299.44,
                    "end": 299.6,
                    "word": " the",
                    "probability": 0.81298828125
                },
                {
                    "start": 299.6,
                    "end": 299.84,
                    "word": " country",
                    "probability": 0.430908203125
                },
                {
                    "start": 299.84,
                    "end": 300.0,
                    "word": " of",
                    "probability": 0.9033203125
                },
                {
                    "start": 300.0,
                    "end": 300.3,
                    "word": " origin",
                    "probability": 0.80810546875
                },
                {
                    "start": 300.3,
                    "end": 300.52,
                    "word": " of",
                    "probability": 0.90478515625
                },
                {
                    "start": 300.52,
                    "end": 300.68,
                    "word": " the",
                    "probability": 0.77685546875
                },
                {
                    "start": 300.68,
                    "end": 301.14,
                    "word": " participants",
                    "probability": 0.89404296875
                },
                {
                    "start": 301.14,
                    "end": 301.52,
                    "word": " on",
                    "probability": 0.90673828125
                },
                {
                    "start": 301.52,
                    "end": 301.7,
                    "word": " day",
                    "probability": 0.79052734375
                },
                {
                    "start": 301.7,
                    "end": 302.02,
                    "word": " one",
                    "probability": 0.6162109375
                },
                {
                    "start": 302.02,
                    "end": 302.58,
                    "word": ".",
                    "probability": 0.951171875
                }
            ]
        },
        {
            "id": 44,
            "seek": 32304,
            "start": 302.58,
            "end": 308.48,
            "text": " It's aggregated, so it does not reveal anything about any of the individuals in the room, right?",
            "tokens": [
                50364,
                467,
                311,
                16743,
                770,
                11,
                370,
                309,
                775,
                406,
                10658,
                1340,
                466,
                604,
                295,
                264,
                5346,
                294,
                264,
                1808,
                11,
                558,
                30,
                50614
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1956168869873146,
            "compression_ratio": 1.5141509433962264,
            "no_speech_prob": 0.0005612373352050781,
            "words": [
                {
                    "start": 302.58,
                    "end": 302.84,
                    "word": " It's",
                    "probability": 0.75830078125
                },
                {
                    "start": 302.84,
                    "end": 303.36,
                    "word": " aggregated",
                    "probability": 0.817138671875
                },
                {
                    "start": 303.36,
                    "end": 303.48,
                    "word": ",",
                    "probability": 0.330078125
                },
                {
                    "start": 303.48,
                    "end": 303.54,
                    "word": " so",
                    "probability": 0.87939453125
                },
                {
                    "start": 303.54,
                    "end": 303.64,
                    "word": " it",
                    "probability": 0.90087890625
                },
                {
                    "start": 303.64,
                    "end": 303.78,
                    "word": " does",
                    "probability": 0.8896484375
                },
                {
                    "start": 303.78,
                    "end": 303.96,
                    "word": " not",
                    "probability": 0.90625
                },
                {
                    "start": 303.96,
                    "end": 304.28,
                    "word": " reveal",
                    "probability": 0.8720703125
                },
                {
                    "start": 304.28,
                    "end": 304.96,
                    "word": " anything",
                    "probability": 0.8330078125
                },
                {
                    "start": 304.96,
                    "end": 305.46,
                    "word": " about",
                    "probability": 0.80712890625
                },
                {
                    "start": 305.46,
                    "end": 305.78,
                    "word": " any",
                    "probability": 0.7861328125
                },
                {
                    "start": 305.78,
                    "end": 305.92,
                    "word": " of",
                    "probability": 0.8681640625
                },
                {
                    "start": 305.92,
                    "end": 306.04,
                    "word": " the",
                    "probability": 0.8046875
                },
                {
                    "start": 306.04,
                    "end": 306.44,
                    "word": " individuals",
                    "probability": 0.66650390625
                },
                {
                    "start": 306.44,
                    "end": 306.66,
                    "word": " in",
                    "probability": 0.89892578125
                },
                {
                    "start": 306.66,
                    "end": 306.8,
                    "word": " the",
                    "probability": 0.81494140625
                },
                {
                    "start": 306.8,
                    "end": 307.02,
                    "word": " room",
                    "probability": 0.81640625
                },
                {
                    "start": 307.02,
                    "end": 307.38,
                    "word": ",",
                    "probability": 0.38916015625
                },
                {
                    "start": 307.38,
                    "end": 307.92,
                    "word": " right",
                    "probability": 0.85546875
                },
                {
                    "start": 307.92,
                    "end": 308.48,
                    "word": "?",
                    "probability": 0.93310546875
                }
            ]
        },
        {
            "id": 45,
            "seek": 32304,
            "start": 308.7,
            "end": 317.16,
            "text": " And it looks like the top country of origin is Italy, whereas Germany has the least participants and France and Spain are in the middle.",
            "tokens": [
                50664,
                400,
                309,
                1542,
                411,
                264,
                1192,
                1941,
                295,
                4957,
                307,
                10705,
                11,
                9735,
                7244,
                575,
                264,
                1935,
                10503,
                293,
                6190,
                293,
                12838,
                366,
                294,
                264,
                2808,
                13,
                51064
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1956168869873146,
            "compression_ratio": 1.5141509433962264,
            "no_speech_prob": 0.0005612373352050781,
            "words": [
                {
                    "start": 308.7,
                    "end": 309.26,
                    "word": " And",
                    "probability": 0.67822265625
                },
                {
                    "start": 309.26,
                    "end": 309.38,
                    "word": " it",
                    "probability": 0.88232421875
                },
                {
                    "start": 309.38,
                    "end": 309.58,
                    "word": " looks",
                    "probability": 0.7958984375
                },
                {
                    "start": 309.58,
                    "end": 309.74,
                    "word": " like",
                    "probability": 0.86767578125
                },
                {
                    "start": 309.74,
                    "end": 309.84,
                    "word": " the",
                    "probability": 0.796875
                },
                {
                    "start": 309.84,
                    "end": 310.02,
                    "word": " top",
                    "probability": 0.81640625
                },
                {
                    "start": 310.02,
                    "end": 310.38,
                    "word": " country",
                    "probability": 0.8193359375
                },
                {
                    "start": 310.38,
                    "end": 310.52,
                    "word": " of",
                    "probability": 0.9033203125
                },
                {
                    "start": 310.52,
                    "end": 310.88,
                    "word": " origin",
                    "probability": 0.81494140625
                },
                {
                    "start": 310.88,
                    "end": 311.22,
                    "word": " is",
                    "probability": 0.90234375
                },
                {
                    "start": 311.22,
                    "end": 311.62,
                    "word": " Italy",
                    "probability": 0.77978515625
                },
                {
                    "start": 311.62,
                    "end": 311.8,
                    "word": ",",
                    "probability": 0.78564453125
                },
                {
                    "start": 311.8,
                    "end": 312.08,
                    "word": " whereas",
                    "probability": 0.666015625
                },
                {
                    "start": 312.08,
                    "end": 312.54,
                    "word": " Germany",
                    "probability": 0.916015625
                },
                {
                    "start": 312.54,
                    "end": 313.14,
                    "word": " has",
                    "probability": 0.888671875
                },
                {
                    "start": 313.14,
                    "end": 313.3,
                    "word": " the",
                    "probability": 0.73876953125
                },
                {
                    "start": 313.3,
                    "end": 313.42,
                    "word": " least",
                    "probability": 0.8662109375
                },
                {
                    "start": 313.42,
                    "end": 313.96,
                    "word": " participants",
                    "probability": 0.88525390625
                },
                {
                    "start": 313.96,
                    "end": 314.34,
                    "word": " and",
                    "probability": 0.53466796875
                },
                {
                    "start": 314.34,
                    "end": 314.86,
                    "word": " France",
                    "probability": 0.87890625
                },
                {
                    "start": 314.86,
                    "end": 315.14,
                    "word": " and",
                    "probability": 0.892578125
                },
                {
                    "start": 315.14,
                    "end": 315.56,
                    "word": " Spain",
                    "probability": 0.90576171875
                },
                {
                    "start": 315.56,
                    "end": 315.98,
                    "word": " are",
                    "probability": 0.89599609375
                },
                {
                    "start": 315.98,
                    "end": 316.24,
                    "word": " in",
                    "probability": 0.90185546875
                },
                {
                    "start": 316.24,
                    "end": 316.38,
                    "word": " the",
                    "probability": 0.8134765625
                },
                {
                    "start": 316.38,
                    "end": 316.6,
                    "word": " middle",
                    "probability": 0.92138671875
                },
                {
                    "start": 316.6,
                    "end": 317.16,
                    "word": ".",
                    "probability": 0.9677734375
                }
            ]
        },
        {
            "id": 46,
            "seek": 32304,
            "start": 317.76,
            "end": 325.08,
            "text": " On day two, the moderator shows the stats for the day. They look pretty similar, right?",
            "tokens": [
                51114,
                1282,
                786,
                732,
                11,
                264,
                37778,
                3110,
                264,
                18152,
                337,
                264,
                786,
                13,
                814,
                574,
                1238,
                2531,
                11,
                558,
                30,
                51464
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1956168869873146,
            "compression_ratio": 1.5141509433962264,
            "no_speech_prob": 0.0005612373352050781,
            "words": [
                {
                    "start": 317.76,
                    "end": 318.32,
                    "word": " On",
                    "probability": 0.89990234375
                },
                {
                    "start": 318.32,
                    "end": 318.54,
                    "word": " day",
                    "probability": 0.732421875
                },
                {
                    "start": 318.54,
                    "end": 318.86,
                    "word": " two",
                    "probability": 0.64794921875
                },
                {
                    "start": 318.86,
                    "end": 319.1,
                    "word": ",",
                    "probability": 0.92236328125
                },
                {
                    "start": 319.1,
                    "end": 319.32,
                    "word": " the",
                    "probability": 0.80859375
                },
                {
                    "start": 319.32,
                    "end": 319.88,
                    "word": " moderator",
                    "probability": 0.8291015625
                },
                {
                    "start": 319.88,
                    "end": 320.52,
                    "word": " shows",
                    "probability": 0.80322265625
                },
                {
                    "start": 320.52,
                    "end": 320.72,
                    "word": " the",
                    "probability": 0.810546875
                },
                {
                    "start": 320.72,
                    "end": 321.02,
                    "word": " stats",
                    "probability": 0.86083984375
                },
                {
                    "start": 321.02,
                    "end": 321.28,
                    "word": " for",
                    "probability": 0.89697265625
                },
                {
                    "start": 321.28,
                    "end": 321.62,
                    "word": " the",
                    "probability": 0.81689453125
                },
                {
                    "start": 321.62,
                    "end": 321.9,
                    "word": " day",
                    "probability": 0.89892578125
                },
                {
                    "start": 321.9,
                    "end": 322.46,
                    "word": ".",
                    "probability": 0.923828125
                },
                {
                    "start": 322.56,
                    "end": 322.74,
                    "word": " They",
                    "probability": 0.70166015625
                },
                {
                    "start": 322.74,
                    "end": 322.92,
                    "word": " look",
                    "probability": 0.90234375
                },
                {
                    "start": 322.92,
                    "end": 323.18,
                    "word": " pretty",
                    "probability": 0.8818359375
                },
                {
                    "start": 323.18,
                    "end": 323.74,
                    "word": " similar",
                    "probability": 0.9189453125
                },
                {
                    "start": 323.74,
                    "end": 324.18,
                    "word": ",",
                    "probability": 0.94140625
                },
                {
                    "start": 324.18,
                    "end": 324.52,
                    "word": " right",
                    "probability": 0.87353515625
                },
                {
                    "start": 324.52,
                    "end": 325.08,
                    "word": "?",
                    "probability": 0.96533203125
                }
            ]
        },
        {
            "id": 47,
            "seek": 35174,
            "start": 325.08,
            "end": 333.38,
            "text": " Except, someone has been paying attention and noticed that there was only one new person on day two.",
            "tokens": [
                50364,
                16192,
                11,
                1580,
                575,
                668,
                6229,
                3202,
                293,
                5694,
                300,
                456,
                390,
                787,
                472,
                777,
                954,
                322,
                786,
                732,
                13,
                50714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2083960843373494,
            "compression_ratio": 1.5358649789029535,
            "no_speech_prob": 0.02716064453125,
            "words": [
                {
                    "start": 325.08,
                    "end": 326.4,
                    "word": " Except",
                    "probability": 0.66064453125
                },
                {
                    "start": 326.4,
                    "end": 326.88,
                    "word": ",",
                    "probability": 0.413330078125
                },
                {
                    "start": 326.88,
                    "end": 327.46,
                    "word": " someone",
                    "probability": 0.85302734375
                },
                {
                    "start": 327.46,
                    "end": 327.72,
                    "word": " has",
                    "probability": 0.88330078125
                },
                {
                    "start": 327.72,
                    "end": 327.9,
                    "word": " been",
                    "probability": 0.814453125
                },
                {
                    "start": 327.9,
                    "end": 328.26,
                    "word": " paying",
                    "probability": 0.85888671875
                },
                {
                    "start": 328.26,
                    "end": 328.72,
                    "word": " attention",
                    "probability": 0.875
                },
                {
                    "start": 328.72,
                    "end": 329.08,
                    "word": " and",
                    "probability": 0.86083984375
                },
                {
                    "start": 329.08,
                    "end": 329.58,
                    "word": " noticed",
                    "probability": 0.79296875
                },
                {
                    "start": 329.58,
                    "end": 330.1,
                    "word": " that",
                    "probability": 0.8369140625
                },
                {
                    "start": 330.1,
                    "end": 330.26,
                    "word": " there",
                    "probability": 0.85302734375
                },
                {
                    "start": 330.26,
                    "end": 330.46,
                    "word": " was",
                    "probability": 0.88623046875
                },
                {
                    "start": 330.46,
                    "end": 330.7,
                    "word": " only",
                    "probability": 0.85205078125
                },
                {
                    "start": 330.7,
                    "end": 331.16,
                    "word": " one",
                    "probability": 0.7880859375
                },
                {
                    "start": 331.16,
                    "end": 331.4,
                    "word": " new",
                    "probability": 0.81005859375
                },
                {
                    "start": 331.4,
                    "end": 331.78,
                    "word": " person",
                    "probability": 0.78759765625
                },
                {
                    "start": 331.78,
                    "end": 332.1,
                    "word": " on",
                    "probability": 0.91259765625
                },
                {
                    "start": 332.1,
                    "end": 332.36,
                    "word": " day",
                    "probability": 0.8095703125
                },
                {
                    "start": 332.36,
                    "end": 332.72,
                    "word": " two",
                    "probability": 0.541015625
                },
                {
                    "start": 332.72,
                    "end": 333.38,
                    "word": ".",
                    "probability": 0.794921875
                }
            ]
        },
        {
            "id": 48,
            "seek": 35174,
            "start": 333.7,
            "end": 337.96,
            "text": " And when they look at the data from both days together, look at that.",
            "tokens": [
                50764,
                400,
                562,
                436,
                574,
                412,
                264,
                1412,
                490,
                1293,
                1708,
                1214,
                11,
                574,
                412,
                300,
                13,
                50964
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2083960843373494,
            "compression_ratio": 1.5358649789029535,
            "no_speech_prob": 0.02716064453125,
            "words": [
                {
                    "start": 333.7,
                    "end": 333.82,
                    "word": " And",
                    "probability": 0.8310546875
                },
                {
                    "start": 333.82,
                    "end": 333.92,
                    "word": " when",
                    "probability": 0.85595703125
                },
                {
                    "start": 333.92,
                    "end": 334.06,
                    "word": " they",
                    "probability": 0.58251953125
                },
                {
                    "start": 334.06,
                    "end": 334.26,
                    "word": " look",
                    "probability": 0.85302734375
                },
                {
                    "start": 334.26,
                    "end": 334.38,
                    "word": " at",
                    "probability": 0.89990234375
                },
                {
                    "start": 334.38,
                    "end": 334.5,
                    "word": " the",
                    "probability": 0.7890625
                },
                {
                    "start": 334.5,
                    "end": 334.72,
                    "word": " data",
                    "probability": 0.5478515625
                },
                {
                    "start": 334.72,
                    "end": 334.94,
                    "word": " from",
                    "probability": 0.7880859375
                },
                {
                    "start": 334.94,
                    "end": 335.24,
                    "word": " both",
                    "probability": 0.81201171875
                },
                {
                    "start": 335.24,
                    "end": 335.54,
                    "word": " days",
                    "probability": 0.740234375
                },
                {
                    "start": 335.54,
                    "end": 335.92,
                    "word": " together",
                    "probability": 0.74560546875
                },
                {
                    "start": 335.92,
                    "end": 336.44,
                    "word": ",",
                    "probability": 0.89013671875
                },
                {
                    "start": 336.44,
                    "end": 336.86,
                    "word": " look",
                    "probability": 0.89990234375
                },
                {
                    "start": 336.86,
                    "end": 337.02,
                    "word": " at",
                    "probability": 0.90673828125
                },
                {
                    "start": 337.02,
                    "end": 337.3,
                    "word": " that",
                    "probability": 0.8828125
                },
                {
                    "start": 337.3,
                    "end": 337.96,
                    "word": ".",
                    "probability": 0.85888671875
                }
            ]
        },
        {
            "id": 49,
            "seek": 35174,
            "start": 338.94,
            "end": 343.16,
            "text": " Can you tell where the specific individual is coming from? Yes.",
            "tokens": [
                51014,
                1664,
                291,
                980,
                689,
                264,
                2685,
                2609,
                307,
                1348,
                490,
                30,
                1079,
                13,
                51214
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2083960843373494,
            "compression_ratio": 1.5358649789029535,
            "no_speech_prob": 0.02716064453125,
            "words": [
                {
                    "start": 338.94,
                    "end": 339.2,
                    "word": " Can",
                    "probability": 0.80029296875
                },
                {
                    "start": 339.2,
                    "end": 339.26,
                    "word": " you",
                    "probability": 0.89453125
                },
                {
                    "start": 339.26,
                    "end": 339.42,
                    "word": " tell",
                    "probability": 0.81787109375
                },
                {
                    "start": 339.42,
                    "end": 339.7,
                    "word": " where",
                    "probability": 0.8994140625
                },
                {
                    "start": 339.7,
                    "end": 339.86,
                    "word": " the",
                    "probability": 0.34423828125
                },
                {
                    "start": 339.86,
                    "end": 340.28,
                    "word": " specific",
                    "probability": 0.802734375
                },
                {
                    "start": 340.28,
                    "end": 340.76,
                    "word": " individual",
                    "probability": 0.7646484375
                },
                {
                    "start": 340.76,
                    "end": 341.02,
                    "word": " is",
                    "probability": 0.90087890625
                },
                {
                    "start": 341.02,
                    "end": 341.28,
                    "word": " coming",
                    "probability": 0.802734375
                },
                {
                    "start": 341.28,
                    "end": 341.58,
                    "word": " from",
                    "probability": 0.81982421875
                },
                {
                    "start": 341.58,
                    "end": 342.16,
                    "word": "?",
                    "probability": 0.9912109375
                },
                {
                    "start": 342.16,
                    "end": 342.5,
                    "word": " Yes",
                    "probability": 0.8955078125
                },
                {
                    "start": 342.5,
                    "end": 343.16,
                    "word": ".",
                    "probability": 0.459228515625
                }
            ]
        },
        {
            "id": 50,
            "seek": 35174,
            "start": 343.42,
            "end": 353.78,
            "text": " Germany. This is how, with two aggregated datasets, the moderator has managed to leak information about a particular participant.",
            "tokens": [
                51264,
                7244,
                13,
                639,
                307,
                577,
                11,
                365,
                732,
                16743,
                770,
                42856,
                11,
                264,
                37778,
                575,
                6453,
                281,
                17143,
                1589,
                466,
                257,
                1729,
                24950,
                13,
                51764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2083960843373494,
            "compression_ratio": 1.5358649789029535,
            "no_speech_prob": 0.02716064453125,
            "words": [
                {
                    "start": 343.42,
                    "end": 343.88,
                    "word": " Germany",
                    "probability": 0.9013671875
                },
                {
                    "start": 343.88,
                    "end": 344.54,
                    "word": ".",
                    "probability": 0.85498046875
                },
                {
                    "start": 344.68,
                    "end": 345.08,
                    "word": " This",
                    "probability": 0.8203125
                },
                {
                    "start": 345.08,
                    "end": 345.26,
                    "word": " is",
                    "probability": 0.90478515625
                },
                {
                    "start": 345.26,
                    "end": 345.6,
                    "word": " how",
                    "probability": 0.89111328125
                },
                {
                    "start": 345.6,
                    "end": 345.84,
                    "word": ",",
                    "probability": 0.42919921875
                },
                {
                    "start": 345.84,
                    "end": 345.92,
                    "word": " with",
                    "probability": 0.81005859375
                },
                {
                    "start": 345.92,
                    "end": 346.12,
                    "word": " two",
                    "probability": 0.79150390625
                },
                {
                    "start": 346.12,
                    "end": 346.82,
                    "word": " aggregated",
                    "probability": 0.8486328125
                },
                {
                    "start": 346.82,
                    "end": 347.34,
                    "word": " datasets",
                    "probability": 0.426513671875
                },
                {
                    "start": 347.34,
                    "end": 347.8,
                    "word": ",",
                    "probability": 0.96044921875
                },
                {
                    "start": 347.8,
                    "end": 347.94,
                    "word": " the",
                    "probability": 0.80615234375
                },
                {
                    "start": 347.94,
                    "end": 348.9,
                    "word": " moderator",
                    "probability": 0.86865234375
                },
                {
                    "start": 348.9,
                    "end": 349.5,
                    "word": " has",
                    "probability": 0.89306640625
                },
                {
                    "start": 349.5,
                    "end": 349.92,
                    "word": " managed",
                    "probability": 0.8388671875
                },
                {
                    "start": 349.92,
                    "end": 350.22,
                    "word": " to",
                    "probability": 0.90673828125
                },
                {
                    "start": 350.22,
                    "end": 350.42,
                    "word": " leak",
                    "probability": 0.90966796875
                },
                {
                    "start": 350.42,
                    "end": 351.14,
                    "word": " information",
                    "probability": 0.7353515625
                },
                {
                    "start": 351.14,
                    "end": 351.84,
                    "word": " about",
                    "probability": 0.73779296875
                },
                {
                    "start": 352.12,
                    "end": 352.6,
                    "word": " a particular",
                    "probability": 0.80029296875
                },
                {
                    "start": 352.6,
                    "end": 353.2,
                    "word": " participant",
                    "probability": 0.8076171875
                },
                {
                    "start": 353.2,
                    "end": 353.78,
                    "word": ".",
                    "probability": 0.99560546875
                }
            ]
        },
        {
            "id": 51,
            "seek": 37664,
            "start": 354.62,
            "end": 362.76,
            "text": " K-anonymity provides stronger guarantees than simple de-identification, but it is not perfect either.",
            "tokens": [
                50364,
                591,
                12,
                282,
                12732,
                507,
                6417,
                7249,
                32567,
                813,
                2199,
                368,
                12,
                1078,
                3774,
                11,
                457,
                309,
                307,
                406,
                2176,
                2139,
                13,
                50764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19481170311188087,
            "compression_ratio": 1.5302325581395348,
            "no_speech_prob": 0.03302001953125,
            "words": [
                {
                    "start": 354.62,
                    "end": 355.36,
                    "word": " K",
                    "probability": 0.2373046875
                },
                {
                    "start": 355.36,
                    "end": 355.88,
                    "word": "-anonymity",
                    "probability": 0.715087890625
                },
                {
                    "start": 355.88,
                    "end": 356.44,
                    "word": " provides",
                    "probability": 0.79296875
                },
                {
                    "start": 356.44,
                    "end": 357.0,
                    "word": " stronger",
                    "probability": 0.7578125
                },
                {
                    "start": 357.0,
                    "end": 357.62,
                    "word": " guarantees",
                    "probability": 0.8017578125
                },
                {
                    "start": 357.62,
                    "end": 357.96,
                    "word": " than",
                    "probability": 0.86669921875
                },
                {
                    "start": 357.96,
                    "end": 358.48,
                    "word": " simple",
                    "probability": 0.8369140625
                },
                {
                    "start": 358.76,
                    "end": 359.5,
                    "word": " de-identification",
                    "probability": 0.90380859375
                },
                {
                    "start": 359.5,
                    "end": 360.06,
                    "word": ",",
                    "probability": 0.64501953125
                },
                {
                    "start": 360.06,
                    "end": 360.28,
                    "word": " but",
                    "probability": 0.87158203125
                },
                {
                    "start": 360.28,
                    "end": 360.42,
                    "word": " it",
                    "probability": 0.87255859375
                },
                {
                    "start": 360.42,
                    "end": 360.6,
                    "word": " is",
                    "probability": 0.86376953125
                },
                {
                    "start": 360.6,
                    "end": 360.88,
                    "word": " not",
                    "probability": 0.89990234375
                },
                {
                    "start": 360.88,
                    "end": 361.58,
                    "word": " perfect",
                    "probability": 0.8515625
                },
                {
                    "start": 361.58,
                    "end": 362.16,
                    "word": " either",
                    "probability": 0.7705078125
                },
                {
                    "start": 362.16,
                    "end": 362.76,
                    "word": ".",
                    "probability": 0.9736328125
                }
            ]
        },
        {
            "id": 52,
            "seek": 37664,
            "start": 362.76,
            "end": 372.76,
            "text": " For example, it is problematic if I'm part of a bucket that is associated with a sensitive attribute, say a particular medical condition.",
            "tokens": [
                50814,
                1171,
                1365,
                11,
                309,
                307,
                19011,
                498,
                286,
                478,
                644,
                295,
                257,
                13058,
                300,
                307,
                6615,
                365,
                257,
                9477,
                19667,
                11,
                584,
                257,
                1729,
                4625,
                4188,
                13,
                51264
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19481170311188087,
            "compression_ratio": 1.5302325581395348,
            "no_speech_prob": 0.03302001953125,
            "words": [
                {
                    "start": 362.76,
                    "end": 363.26,
                    "word": " For",
                    "probability": 0.9140625
                },
                {
                    "start": 363.26,
                    "end": 363.66,
                    "word": " example",
                    "probability": 0.9345703125
                },
                {
                    "start": 363.66,
                    "end": 364.3,
                    "word": ",",
                    "probability": 0.98583984375
                },
                {
                    "start": 364.3,
                    "end": 364.5,
                    "word": " it",
                    "probability": 0.89111328125
                },
                {
                    "start": 364.5,
                    "end": 364.66,
                    "word": " is",
                    "probability": 0.89697265625
                },
                {
                    "start": 364.66,
                    "end": 365.12,
                    "word": " problematic",
                    "probability": 0.82373046875
                },
                {
                    "start": 365.12,
                    "end": 365.48,
                    "word": " if",
                    "probability": 0.896484375
                },
                {
                    "start": 365.48,
                    "end": 365.66,
                    "word": " I'm",
                    "probability": 0.90625
                },
                {
                    "start": 365.66,
                    "end": 365.94,
                    "word": " part",
                    "probability": 0.81005859375
                },
                {
                    "start": 365.94,
                    "end": 366.06,
                    "word": " of",
                    "probability": 0.90380859375
                },
                {
                    "start": 366.22,
                    "end": 366.48,
                    "word": " a bucket",
                    "probability": 0.71923828125
                },
                {
                    "start": 366.48,
                    "end": 366.76,
                    "word": " that",
                    "probability": 0.87255859375
                },
                {
                    "start": 366.76,
                    "end": 367.0,
                    "word": " is",
                    "probability": 0.89892578125
                },
                {
                    "start": 367.0,
                    "end": 367.66,
                    "word": " associated",
                    "probability": 0.8544921875
                },
                {
                    "start": 367.66,
                    "end": 368.0,
                    "word": " with",
                    "probability": 0.8115234375
                },
                {
                    "start": 368.24,
                    "end": 368.58,
                    "word": " a sensitive",
                    "probability": 0.80908203125
                },
                {
                    "start": 368.58,
                    "end": 369.1,
                    "word": " attribute",
                    "probability": 0.5771484375
                },
                {
                    "start": 369.1,
                    "end": 369.78,
                    "word": ",",
                    "probability": 0.87841796875
                },
                {
                    "start": 369.78,
                    "end": 370.1,
                    "word": " say",
                    "probability": 0.88525390625
                },
                {
                    "start": 370.42,
                    "end": 370.86,
                    "word": " a particular",
                    "probability": 0.8447265625
                },
                {
                    "start": 370.86,
                    "end": 371.44,
                    "word": " medical",
                    "probability": 0.92333984375
                },
                {
                    "start": 371.44,
                    "end": 372.02,
                    "word": " condition",
                    "probability": 0.83642578125
                },
                {
                    "start": 372.02,
                    "end": 372.76,
                    "word": ".",
                    "probability": 0.91796875
                }
            ]
        },
        {
            "id": 53,
            "seek": 37664,
            "start": 372.86,
            "end": 378.68,
            "text": " It's not helpful that I'm indistinguishable from K other users having the same condition.",
            "tokens": [
                51314,
                467,
                311,
                406,
                4961,
                300,
                286,
                478,
                1016,
                468,
                7050,
                742,
                712,
                490,
                591,
                661,
                5022,
                1419,
                264,
                912,
                4188,
                13,
                51564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19481170311188087,
            "compression_ratio": 1.5302325581395348,
            "no_speech_prob": 0.03302001953125,
            "words": [
                {
                    "start": 372.86,
                    "end": 373.22,
                    "word": " It's",
                    "probability": 0.93017578125
                },
                {
                    "start": 373.22,
                    "end": 373.38,
                    "word": " not",
                    "probability": 0.455322265625
                },
                {
                    "start": 373.38,
                    "end": 373.78,
                    "word": " helpful",
                    "probability": 0.82421875
                },
                {
                    "start": 373.78,
                    "end": 374.04,
                    "word": " that",
                    "probability": 0.8662109375
                },
                {
                    "start": 374.04,
                    "end": 374.24,
                    "word": " I'm",
                    "probability": 0.966552734375
                },
                {
                    "start": 374.24,
                    "end": 375.06,
                    "word": " indistinguishable",
                    "probability": 0.96201171875
                },
                {
                    "start": 375.06,
                    "end": 375.34,
                    "word": " from",
                    "probability": 0.802734375
                },
                {
                    "start": 375.34,
                    "end": 375.68,
                    "word": " K",
                    "probability": 0.7109375
                },
                {
                    "start": 375.68,
                    "end": 375.9,
                    "word": " other",
                    "probability": 0.48486328125
                },
                {
                    "start": 375.9,
                    "end": 376.34,
                    "word": " users",
                    "probability": 0.8369140625
                },
                {
                    "start": 376.34,
                    "end": 376.74,
                    "word": " having",
                    "probability": 0.7861328125
                },
                {
                    "start": 376.74,
                    "end": 376.98,
                    "word": " the",
                    "probability": 0.802734375
                },
                {
                    "start": 376.98,
                    "end": 377.36,
                    "word": " same",
                    "probability": 0.81103515625
                },
                {
                    "start": 377.36,
                    "end": 377.94,
                    "word": " condition",
                    "probability": 0.853515625
                },
                {
                    "start": 377.94,
                    "end": 378.68,
                    "word": ".",
                    "probability": 0.9814453125
                }
            ]
        },
        {
            "id": 54,
            "seek": 39618,
            "start": 378.68,
            "end": 388.12,
            "text": " While I cannot be associated with any particular record in the database, my sensitive attribute in that example is revealed.",
            "tokens": [
                50364,
                3987,
                286,
                2644,
                312,
                6615,
                365,
                604,
                1729,
                2136,
                294,
                264,
                8149,
                11,
                452,
                9477,
                19667,
                294,
                300,
                1365,
                307,
                9599,
                13,
                50764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19889322916666666,
            "compression_ratio": 1.4230769230769231,
            "no_speech_prob": 0.0025501251220703125,
            "words": [
                {
                    "start": 378.68,
                    "end": 379.04,
                    "word": " While",
                    "probability": 0.77685546875
                },
                {
                    "start": 379.04,
                    "end": 379.2,
                    "word": " I",
                    "probability": 0.92041015625
                },
                {
                    "start": 379.2,
                    "end": 379.52,
                    "word": " cannot",
                    "probability": 0.6962890625
                },
                {
                    "start": 379.52,
                    "end": 379.98,
                    "word": " be",
                    "probability": 0.89892578125
                },
                {
                    "start": 379.98,
                    "end": 380.58,
                    "word": " associated",
                    "probability": 0.85595703125
                },
                {
                    "start": 380.58,
                    "end": 380.9,
                    "word": " with",
                    "probability": 0.814453125
                },
                {
                    "start": 380.9,
                    "end": 381.2,
                    "word": " any",
                    "probability": 0.83251953125
                },
                {
                    "start": 381.2,
                    "end": 381.66,
                    "word": " particular",
                    "probability": 0.7958984375
                },
                {
                    "start": 381.66,
                    "end": 382.18,
                    "word": " record",
                    "probability": 0.91455078125
                },
                {
                    "start": 382.18,
                    "end": 382.46,
                    "word": " in",
                    "probability": 0.90234375
                },
                {
                    "start": 382.46,
                    "end": 382.64,
                    "word": " the",
                    "probability": 0.80859375
                },
                {
                    "start": 382.64,
                    "end": 383.22,
                    "word": " database",
                    "probability": 0.7998046875
                },
                {
                    "start": 383.22,
                    "end": 383.6,
                    "word": ",",
                    "probability": 0.8876953125
                },
                {
                    "start": 383.6,
                    "end": 383.86,
                    "word": " my",
                    "probability": 0.88232421875
                },
                {
                    "start": 383.86,
                    "end": 384.3,
                    "word": " sensitive",
                    "probability": 0.75732421875
                },
                {
                    "start": 384.3,
                    "end": 385.06,
                    "word": " attribute",
                    "probability": 0.47265625
                },
                {
                    "start": 385.06,
                    "end": 385.46,
                    "word": " in",
                    "probability": 0.8837890625
                },
                {
                    "start": 385.46,
                    "end": 385.68,
                    "word": " that",
                    "probability": 0.8876953125
                },
                {
                    "start": 385.68,
                    "end": 386.16,
                    "word": " example",
                    "probability": 0.92236328125
                },
                {
                    "start": 386.16,
                    "end": 386.66,
                    "word": " is",
                    "probability": 0.90234375
                },
                {
                    "start": 386.66,
                    "end": 387.22,
                    "word": " revealed",
                    "probability": 0.89306640625
                },
                {
                    "start": 387.22,
                    "end": 388.12,
                    "word": ".",
                    "probability": 0.974609375
                }
            ]
        },
        {
            "id": 55,
            "seek": 39618,
            "start": 388.12,
            "end": 398.22,
            "text": " Here is another example. Imagine two aggregate buckets, 99 users plus me and 99 users without me.",
            "tokens": [
                50814,
                1692,
                307,
                1071,
                1365,
                13,
                11739,
                732,
                26118,
                32191,
                11,
                11803,
                5022,
                1804,
                385,
                293,
                11803,
                5022,
                1553,
                385,
                13,
                51264
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19889322916666666,
            "compression_ratio": 1.4230769230769231,
            "no_speech_prob": 0.0025501251220703125,
            "words": [
                {
                    "start": 388.12,
                    "end": 388.46,
                    "word": " Here",
                    "probability": 0.83154296875
                },
                {
                    "start": 388.46,
                    "end": 388.58,
                    "word": " is",
                    "probability": 0.591796875
                },
                {
                    "start": 388.58,
                    "end": 388.86,
                    "word": " another",
                    "probability": 0.82177734375
                },
                {
                    "start": 388.86,
                    "end": 389.3,
                    "word": " example",
                    "probability": 0.9296875
                },
                {
                    "start": 389.3,
                    "end": 390.08,
                    "word": ".",
                    "probability": 0.9033203125
                },
                {
                    "start": 390.08,
                    "end": 390.52,
                    "word": " Imagine",
                    "probability": 0.80908203125
                },
                {
                    "start": 390.52,
                    "end": 391.14,
                    "word": " two",
                    "probability": 0.75341796875
                },
                {
                    "start": 391.14,
                    "end": 391.7,
                    "word": " aggregate",
                    "probability": 0.459716796875
                },
                {
                    "start": 391.7,
                    "end": 392.16,
                    "word": " buckets",
                    "probability": 0.7099609375
                },
                {
                    "start": 392.16,
                    "end": 392.78,
                    "word": ",",
                    "probability": 0.58984375
                },
                {
                    "start": 392.78,
                    "end": 393.16,
                    "word": " 99",
                    "probability": 0.859375
                },
                {
                    "start": 393.16,
                    "end": 393.64,
                    "word": " users",
                    "probability": 0.654296875
                },
                {
                    "start": 393.64,
                    "end": 394.06,
                    "word": " plus",
                    "probability": 0.83447265625
                },
                {
                    "start": 394.06,
                    "end": 394.4,
                    "word": " me",
                    "probability": 0.90478515625
                },
                {
                    "start": 394.4,
                    "end": 395.08,
                    "word": " and",
                    "probability": 0.6083984375
                },
                {
                    "start": 395.08,
                    "end": 395.56,
                    "word": " 99",
                    "probability": 0.8984375
                },
                {
                    "start": 395.56,
                    "end": 396.16,
                    "word": " users",
                    "probability": 0.82763671875
                },
                {
                    "start": 396.16,
                    "end": 396.94,
                    "word": " without",
                    "probability": 0.80859375
                },
                {
                    "start": 396.94,
                    "end": 397.32,
                    "word": " me",
                    "probability": 0.91162109375
                },
                {
                    "start": 397.32,
                    "end": 398.22,
                    "word": ".",
                    "probability": 0.97021484375
                }
            ]
        },
        {
            "id": 56,
            "seek": 42140,
            "start": 398.22,
            "end": 410.6,
            "text": " Publishing aggregate information about those two buckets is K-anonymous, but it's possible to identify me and my sensitive attributes by differencing those two buckets.",
            "tokens": [
                50364,
                21808,
                18533,
                26118,
                1589,
                466,
                729,
                732,
                32191,
                307,
                591,
                12,
                282,
                18092,
                11,
                457,
                309,
                311,
                1944,
                281,
                5876,
                385,
                293,
                452,
                9477,
                17212,
                538,
                743,
                13644,
                729,
                732,
                32191,
                13,
                50914
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1878210616438356,
            "compression_ratio": 1.5414634146341464,
            "no_speech_prob": 0.00818634033203125,
            "words": [
                {
                    "start": 398.22,
                    "end": 398.68,
                    "word": " Publishing",
                    "probability": 0.736083984375
                },
                {
                    "start": 398.68,
                    "end": 399.26,
                    "word": " aggregate",
                    "probability": 0.70751953125
                },
                {
                    "start": 399.26,
                    "end": 399.96,
                    "word": " information",
                    "probability": 0.69775390625
                },
                {
                    "start": 399.96,
                    "end": 400.36,
                    "word": " about",
                    "probability": 0.47509765625
                },
                {
                    "start": 400.36,
                    "end": 400.66,
                    "word": " those",
                    "probability": 0.84912109375
                },
                {
                    "start": 400.66,
                    "end": 400.96,
                    "word": " two",
                    "probability": 0.806640625
                },
                {
                    "start": 400.96,
                    "end": 401.36,
                    "word": " buckets",
                    "probability": 0.71337890625
                },
                {
                    "start": 401.36,
                    "end": 401.7,
                    "word": " is",
                    "probability": 0.900390625
                },
                {
                    "start": 401.7,
                    "end": 402.0,
                    "word": " K",
                    "probability": 0.06671142578125
                },
                {
                    "start": 402.0,
                    "end": 402.54,
                    "word": "-anonymous",
                    "probability": 0.68505859375
                },
                {
                    "start": 402.54,
                    "end": 403.22,
                    "word": ",",
                    "probability": 0.49169921875
                },
                {
                    "start": 403.22,
                    "end": 403.36,
                    "word": " but",
                    "probability": 0.8798828125
                },
                {
                    "start": 403.36,
                    "end": 403.54,
                    "word": " it's",
                    "probability": 0.934814453125
                },
                {
                    "start": 403.54,
                    "end": 403.96,
                    "word": " possible",
                    "probability": 0.853515625
                },
                {
                    "start": 403.96,
                    "end": 404.18,
                    "word": " to",
                    "probability": 0.9033203125
                },
                {
                    "start": 404.18,
                    "end": 404.7,
                    "word": " identify",
                    "probability": 0.90673828125
                },
                {
                    "start": 404.7,
                    "end": 405.22,
                    "word": " me",
                    "probability": 0.89599609375
                },
                {
                    "start": 405.22,
                    "end": 405.68,
                    "word": " and",
                    "probability": 0.88330078125
                },
                {
                    "start": 405.68,
                    "end": 405.88,
                    "word": " my",
                    "probability": 0.90576171875
                },
                {
                    "start": 405.88,
                    "end": 406.28,
                    "word": " sensitive",
                    "probability": 0.71142578125
                },
                {
                    "start": 406.28,
                    "end": 407.26,
                    "word": " attributes",
                    "probability": 0.64306640625
                },
                {
                    "start": 407.26,
                    "end": 407.82,
                    "word": " by",
                    "probability": 0.9033203125
                },
                {
                    "start": 407.82,
                    "end": 408.66,
                    "word": " differencing",
                    "probability": 0.6741943359375
                },
                {
                    "start": 408.66,
                    "end": 409.18,
                    "word": " those",
                    "probability": 0.8515625
                },
                {
                    "start": 409.18,
                    "end": 409.52,
                    "word": " two",
                    "probability": 0.8134765625
                },
                {
                    "start": 409.52,
                    "end": 409.94,
                    "word": " buckets",
                    "probability": 0.7001953125
                },
                {
                    "start": 409.94,
                    "end": 410.6,
                    "word": ".",
                    "probability": 0.990234375
                }
            ]
        },
        {
            "id": 57,
            "seek": 42140,
            "start": 410.6,
            "end": 423.44,
            "text": " At last, if the attacker can modify the database and create fake entries, there's no anonymity for my data if I'm part of a crowd of 99 fake users.",
            "tokens": [
                50964,
                1711,
                1036,
                11,
                498,
                264,
                35871,
                393,
                16927,
                264,
                8149,
                293,
                1884,
                7592,
                23041,
                11,
                456,
                311,
                572,
                37293,
                507,
                337,
                452,
                1412,
                498,
                286,
                478,
                644,
                295,
                257,
                6919,
                295,
                11803,
                7592,
                5022,
                13,
                51564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1878210616438356,
            "compression_ratio": 1.5414634146341464,
            "no_speech_prob": 0.00818634033203125,
            "words": [
                {
                    "start": 410.6,
                    "end": 410.88,
                    "word": " At",
                    "probability": 0.89892578125
                },
                {
                    "start": 410.88,
                    "end": 411.24,
                    "word": " last",
                    "probability": 0.79736328125
                },
                {
                    "start": 411.24,
                    "end": 411.78,
                    "word": ",",
                    "probability": 0.984375
                },
                {
                    "start": 411.78,
                    "end": 411.92,
                    "word": " if",
                    "probability": 0.904296875
                },
                {
                    "start": 411.92,
                    "end": 412.04,
                    "word": " the",
                    "probability": 0.79541015625
                },
                {
                    "start": 412.04,
                    "end": 412.26,
                    "word": " attacker",
                    "probability": 0.73291015625
                },
                {
                    "start": 412.26,
                    "end": 412.54,
                    "word": " can",
                    "probability": 0.9091796875
                },
                {
                    "start": 412.54,
                    "end": 413.06,
                    "word": " modify",
                    "probability": 0.86279296875
                },
                {
                    "start": 413.06,
                    "end": 413.32,
                    "word": " the",
                    "probability": 0.81005859375
                },
                {
                    "start": 413.32,
                    "end": 413.82,
                    "word": " database",
                    "probability": 0.81982421875
                },
                {
                    "start": 413.82,
                    "end": 414.16,
                    "word": " and",
                    "probability": 0.89111328125
                },
                {
                    "start": 414.16,
                    "end": 414.56,
                    "word": " create",
                    "probability": 0.84814453125
                },
                {
                    "start": 414.56,
                    "end": 415.04,
                    "word": " fake",
                    "probability": 0.8564453125
                },
                {
                    "start": 415.04,
                    "end": 415.62,
                    "word": " entries",
                    "probability": 0.78076171875
                },
                {
                    "start": 415.62,
                    "end": 416.08,
                    "word": ",",
                    "probability": 0.97607421875
                },
                {
                    "start": 416.08,
                    "end": 416.36,
                    "word": " there's",
                    "probability": 0.89501953125
                },
                {
                    "start": 416.36,
                    "end": 416.74,
                    "word": " no",
                    "probability": 0.88525390625
                },
                {
                    "start": 416.74,
                    "end": 417.46,
                    "word": " anonymity",
                    "probability": 0.89453125
                },
                {
                    "start": 417.46,
                    "end": 417.76,
                    "word": " for",
                    "probability": 0.83154296875
                },
                {
                    "start": 417.76,
                    "end": 417.98,
                    "word": " my",
                    "probability": 0.8994140625
                },
                {
                    "start": 417.98,
                    "end": 418.34,
                    "word": " data",
                    "probability": 0.822265625
                },
                {
                    "start": 418.34,
                    "end": 418.98,
                    "word": " if",
                    "probability": 0.66845703125
                },
                {
                    "start": 418.98,
                    "end": 419.14,
                    "word": " I'm",
                    "probability": 0.9931640625
                },
                {
                    "start": 419.14,
                    "end": 419.4,
                    "word": " part",
                    "probability": 0.8125
                },
                {
                    "start": 419.4,
                    "end": 419.58,
                    "word": " of",
                    "probability": 0.9072265625
                },
                {
                    "start": 419.72,
                    "end": 420.6,
                    "word": " a crowd",
                    "probability": 0.56787109375
                },
                {
                    "start": 420.6,
                    "end": 420.96,
                    "word": " of",
                    "probability": 0.90771484375
                },
                {
                    "start": 420.96,
                    "end": 421.54,
                    "word": " 99",
                    "probability": 0.94970703125
                },
                {
                    "start": 421.54,
                    "end": 422.1,
                    "word": " fake",
                    "probability": 0.8408203125
                },
                {
                    "start": 422.1,
                    "end": 422.64,
                    "word": " users",
                    "probability": 0.8681640625
                },
                {
                    "start": 422.64,
                    "end": 423.44,
                    "word": ".",
                    "probability": 0.994140625
                }
            ]
        },
        {
            "id": 58,
            "seek": 43761,
            "start": 423.44,
            "end": 430.48,
            "text": " Those type of attacks aren't always feasible or relevant, but they cannot be excluded in general.",
            "tokens": [
                50364,
                3950,
                2010,
                295,
                8122,
                3212,
                380,
                1009,
                26648,
                420,
                7340,
                11,
                457,
                436,
                2644,
                312,
                29486,
                294,
                2674,
                13,
                50664
            ],
            "temperature": 0.0,
            "avg_logprob": -0.22187499602635702,
            "compression_ratio": 1.2867132867132867,
            "no_speech_prob": 0.00394439697265625,
            "words": [
                {
                    "start": 423.44,
                    "end": 423.94,
                    "word": " Those",
                    "probability": 0.80078125
                },
                {
                    "start": 423.94,
                    "end": 424.22,
                    "word": " type",
                    "probability": 0.53271484375
                },
                {
                    "start": 424.22,
                    "end": 424.32,
                    "word": " of",
                    "probability": 0.833984375
                },
                {
                    "start": 424.32,
                    "end": 424.66,
                    "word": " attacks",
                    "probability": 0.71142578125
                },
                {
                    "start": 424.66,
                    "end": 425.06,
                    "word": " aren't",
                    "probability": 0.849365234375
                },
                {
                    "start": 425.06,
                    "end": 425.44,
                    "word": " always",
                    "probability": 0.78369140625
                },
                {
                    "start": 425.44,
                    "end": 425.86,
                    "word": " feasible",
                    "probability": 0.81103515625
                },
                {
                    "start": 425.86,
                    "end": 426.26,
                    "word": " or",
                    "probability": 0.90234375
                },
                {
                    "start": 426.26,
                    "end": 426.9,
                    "word": " relevant",
                    "probability": 0.87060546875
                },
                {
                    "start": 426.9,
                    "end": 427.28,
                    "word": ",",
                    "probability": 0.630859375
                },
                {
                    "start": 427.28,
                    "end": 427.46,
                    "word": " but",
                    "probability": 0.87939453125
                },
                {
                    "start": 427.46,
                    "end": 427.58,
                    "word": " they",
                    "probability": 0.552734375
                },
                {
                    "start": 427.58,
                    "end": 427.96,
                    "word": " cannot",
                    "probability": 0.75537109375
                },
                {
                    "start": 427.96,
                    "end": 428.3,
                    "word": " be",
                    "probability": 0.908203125
                },
                {
                    "start": 428.3,
                    "end": 428.82,
                    "word": " excluded",
                    "probability": 0.76171875
                },
                {
                    "start": 428.82,
                    "end": 429.22,
                    "word": " in",
                    "probability": 0.8955078125
                },
                {
                    "start": 429.22,
                    "end": 429.7,
                    "word": " general",
                    "probability": 0.79638671875
                },
                {
                    "start": 429.7,
                    "end": 430.48,
                    "word": ".",
                    "probability": 0.96337890625
                }
            ]
        },
        {
            "id": 59,
            "seek": 43761,
            "start": 430.62,
            "end": 439.65,
            "text": " Is there a better way to define anonymity? I think so. Let's use differential privacy.",
            "tokens": [
                50714,
                1119,
                456,
                257,
                1101,
                636,
                281,
                6964,
                37293,
                507,
                30,
                286,
                519,
                370,
                13,
                961,
                311,
                764,
                15756,
                11427,
                13,
                51164
            ],
            "temperature": 0.0,
            "avg_logprob": -0.22187499602635702,
            "compression_ratio": 1.2867132867132867,
            "no_speech_prob": 0.00394439697265625,
            "words": [
                {
                    "start": 430.62,
                    "end": 431.0,
                    "word": " Is",
                    "probability": 0.90478515625
                },
                {
                    "start": 431.0,
                    "end": 431.14,
                    "word": " there",
                    "probability": 0.69921875
                },
                {
                    "start": 431.44,
                    "end": 431.78,
                    "word": " a better",
                    "probability": 0.82861328125
                },
                {
                    "start": 431.78,
                    "end": 432.26,
                    "word": " way",
                    "probability": 0.89599609375
                },
                {
                    "start": 432.26,
                    "end": 432.56,
                    "word": " to",
                    "probability": 0.90625
                },
                {
                    "start": 432.56,
                    "end": 432.96,
                    "word": " define",
                    "probability": 0.8876953125
                },
                {
                    "start": 432.96,
                    "end": 433.58,
                    "word": " anonymity",
                    "probability": 0.8515625
                },
                {
                    "start": 433.58,
                    "end": 434.36,
                    "word": "?",
                    "probability": 0.9912109375
                },
                {
                    "start": 435.4,
                    "end": 436.18,
                    "word": " I",
                    "probability": 0.99072265625
                },
                {
                    "start": 436.18,
                    "end": 436.34,
                    "word": " think",
                    "probability": 0.81103515625
                },
                {
                    "start": 436.34,
                    "end": 436.62,
                    "word": " so",
                    "probability": 0.900390625
                },
                {
                    "start": 436.62,
                    "end": 437.34,
                    "word": ".",
                    "probability": 0.89404296875
                },
                {
                    "start": 437.34,
                    "end": 437.68,
                    "word": " Let's",
                    "probability": 0.94287109375
                },
                {
                    "start": 437.68,
                    "end": 438.04,
                    "word": " use",
                    "probability": 0.779296875
                },
                {
                    "start": 438.04,
                    "end": 438.68,
                    "word": " differential",
                    "probability": 0.689453125
                },
                {
                    "start": 438.68,
                    "end": 439.26,
                    "word": " privacy",
                    "probability": 0.65625
                },
                {
                    "start": 439.26,
                    "end": 439.65,
                    "word": ".",
                    "probability": 0.9912109375
                }
            ]
        },
        {
            "id": 60,
            "seek": 46761,
            "start": 440.65,
            "end": 453.69,
            "text": " The moderator would similarly announce that they're showing insights about the audience, but they would like to announce, I'm going to share insights about the audience, randomly changing the numbers just a little.",
            "tokens": [
                50414,
                440,
                37778,
                576,
                14138,
                7478,
                300,
                436,
                434,
                4099,
                14310,
                466,
                264,
                4034,
                11,
                457,
                436,
                576,
                411,
                281,
                7478,
                11,
                286,
                478,
                516,
                281,
                2073,
                14310,
                466,
                264,
                4034,
                11,
                16979,
                4473,
                264,
                3547,
                445,
                257,
                707,
                13,
                51014
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2507267476514328,
            "compression_ratio": 1.7844827586206897,
            "no_speech_prob": 0.403076171875,
            "words": [
                {
                    "start": 440.65,
                    "end": 440.99,
                    "word": " The",
                    "probability": 0.75927734375
                },
                {
                    "start": 440.99,
                    "end": 441.49,
                    "word": " moderator",
                    "probability": 0.84228515625
                },
                {
                    "start": 441.49,
                    "end": 441.91,
                    "word": " would",
                    "probability": 0.78369140625
                },
                {
                    "start": 441.91,
                    "end": 442.41,
                    "word": " similarly",
                    "probability": 0.8671875
                },
                {
                    "start": 442.41,
                    "end": 442.87,
                    "word": " announce",
                    "probability": 0.82861328125
                },
                {
                    "start": 442.87,
                    "end": 443.09,
                    "word": " that",
                    "probability": 0.865234375
                },
                {
                    "start": 443.09,
                    "end": 443.27,
                    "word": " they're",
                    "probability": 0.798095703125
                },
                {
                    "start": 443.27,
                    "end": 443.59,
                    "word": " showing",
                    "probability": 0.791015625
                },
                {
                    "start": 443.59,
                    "end": 444.13,
                    "word": " insights",
                    "probability": 0.8232421875
                },
                {
                    "start": 444.13,
                    "end": 444.43,
                    "word": " about",
                    "probability": 0.81298828125
                },
                {
                    "start": 444.43,
                    "end": 444.65,
                    "word": " the",
                    "probability": 0.81494140625
                },
                {
                    "start": 444.65,
                    "end": 445.07,
                    "word": " audience",
                    "probability": 0.80615234375
                },
                {
                    "start": 445.07,
                    "end": 445.25,
                    "word": ",",
                    "probability": 0.49462890625
                },
                {
                    "start": 445.25,
                    "end": 445.33,
                    "word": " but",
                    "probability": 0.8203125
                },
                {
                    "start": 445.33,
                    "end": 445.43,
                    "word": " they",
                    "probability": 0.329345703125
                },
                {
                    "start": 445.43,
                    "end": 445.61,
                    "word": " would",
                    "probability": 0.8408203125
                },
                {
                    "start": 445.61,
                    "end": 445.81,
                    "word": " like",
                    "probability": 0.8642578125
                },
                {
                    "start": 445.81,
                    "end": 445.97,
                    "word": " to",
                    "probability": 0.90869140625
                },
                {
                    "start": 445.97,
                    "end": 446.35,
                    "word": " announce",
                    "probability": 0.85009765625
                },
                {
                    "start": 446.35,
                    "end": 446.67,
                    "word": ",",
                    "probability": 0.59228515625
                },
                {
                    "start": 446.67,
                    "end": 446.99,
                    "word": " I'm",
                    "probability": 0.746337890625
                },
                {
                    "start": 446.99,
                    "end": 447.21,
                    "word": " going",
                    "probability": 0.85400390625
                },
                {
                    "start": 447.21,
                    "end": 447.35,
                    "word": " to",
                    "probability": 0.90625
                },
                {
                    "start": 447.35,
                    "end": 447.63,
                    "word": " share",
                    "probability": 0.72802734375
                },
                {
                    "start": 447.63,
                    "end": 448.13,
                    "word": " insights",
                    "probability": 0.8349609375
                },
                {
                    "start": 448.13,
                    "end": 448.35,
                    "word": " about",
                    "probability": 0.8056640625
                },
                {
                    "start": 448.35,
                    "end": 448.57,
                    "word": " the",
                    "probability": 0.8134765625
                },
                {
                    "start": 448.57,
                    "end": 448.95,
                    "word": " audience",
                    "probability": 0.8134765625
                },
                {
                    "start": 448.95,
                    "end": 449.39,
                    "word": ",",
                    "probability": 0.6845703125
                },
                {
                    "start": 449.39,
                    "end": 449.83,
                    "word": " randomly",
                    "probability": 0.6357421875
                },
                {
                    "start": 449.83,
                    "end": 450.53,
                    "word": " changing",
                    "probability": 0.79296875
                },
                {
                    "start": 450.53,
                    "end": 451.19,
                    "word": " the",
                    "probability": 0.806640625
                },
                {
                    "start": 451.19,
                    "end": 451.59,
                    "word": " numbers",
                    "probability": 0.857421875
                },
                {
                    "start": 451.59,
                    "end": 452.29,
                    "word": " just",
                    "probability": 0.8017578125
                },
                {
                    "start": 452.75,
                    "end": 453.01,
                    "word": " a little",
                    "probability": 0.80712890625
                },
                {
                    "start": 453.01,
                    "end": 453.69,
                    "word": ".",
                    "probability": 0.9765625
                }
            ]
        },
        {
            "id": 61,
            "seek": 46761,
            "start": 454.07,
            "end": 469.63,
            "text": " Look at the day one stats, how these stats look compared to the real numbers. They're slightly different. So yes, it is less precise, but you still get the same insights about the statistical trends.",
            "tokens": [
                51064,
                2053,
                412,
                264,
                786,
                472,
                18152,
                11,
                577,
                613,
                18152,
                574,
                5347,
                281,
                264,
                957,
                3547,
                13,
                814,
                434,
                4748,
                819,
                13,
                407,
                2086,
                11,
                309,
                307,
                1570,
                13600,
                11,
                457,
                291,
                920,
                483,
                264,
                912,
                14310,
                466,
                264,
                22820,
                13892,
                13,
                51814
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2507267476514328,
            "compression_ratio": 1.7844827586206897,
            "no_speech_prob": 0.403076171875,
            "words": [
                {
                    "start": 454.07,
                    "end": 454.45,
                    "word": " Look",
                    "probability": 0.828125
                },
                {
                    "start": 454.45,
                    "end": 454.55,
                    "word": " at",
                    "probability": 0.89990234375
                },
                {
                    "start": 454.55,
                    "end": 454.69,
                    "word": " the",
                    "probability": 0.65185546875
                },
                {
                    "start": 454.69,
                    "end": 454.87,
                    "word": " day",
                    "probability": 0.76611328125
                },
                {
                    "start": 454.87,
                    "end": 455.21,
                    "word": " one",
                    "probability": 0.7431640625
                },
                {
                    "start": 455.21,
                    "end": 455.65,
                    "word": " stats",
                    "probability": 0.853515625
                },
                {
                    "start": 455.65,
                    "end": 456.51,
                    "word": ",",
                    "probability": 0.52099609375
                },
                {
                    "start": 456.51,
                    "end": 457.55,
                    "word": " how",
                    "probability": 0.873046875
                },
                {
                    "start": 457.55,
                    "end": 458.35,
                    "word": " these",
                    "probability": 0.3759765625
                },
                {
                    "start": 458.35,
                    "end": 458.67,
                    "word": " stats",
                    "probability": 0.8798828125
                },
                {
                    "start": 458.67,
                    "end": 459.33,
                    "word": " look",
                    "probability": 0.8935546875
                },
                {
                    "start": 459.33,
                    "end": 459.89,
                    "word": " compared",
                    "probability": 0.83349609375
                },
                {
                    "start": 459.89,
                    "end": 460.15,
                    "word": " to",
                    "probability": 0.90673828125
                },
                {
                    "start": 460.15,
                    "end": 460.29,
                    "word": " the",
                    "probability": 0.8154296875
                },
                {
                    "start": 460.29,
                    "end": 460.53,
                    "word": " real",
                    "probability": 0.8779296875
                },
                {
                    "start": 460.53,
                    "end": 460.99,
                    "word": " numbers",
                    "probability": 0.87060546875
                },
                {
                    "start": 460.99,
                    "end": 461.51,
                    "word": ".",
                    "probability": 0.787109375
                },
                {
                    "start": 461.51,
                    "end": 461.75,
                    "word": " They're",
                    "probability": 0.8212890625
                },
                {
                    "start": 461.75,
                    "end": 462.13,
                    "word": " slightly",
                    "probability": 0.78955078125
                },
                {
                    "start": 462.13,
                    "end": 462.61,
                    "word": " different",
                    "probability": 0.6904296875
                },
                {
                    "start": 462.61,
                    "end": 463.17,
                    "word": ".",
                    "probability": 0.84814453125
                },
                {
                    "start": 463.17,
                    "end": 463.51,
                    "word": " So",
                    "probability": 0.8935546875
                },
                {
                    "start": 463.51,
                    "end": 463.95,
                    "word": " yes",
                    "probability": 0.72998046875
                },
                {
                    "start": 463.95,
                    "end": 464.23,
                    "word": ",",
                    "probability": 0.97998046875
                },
                {
                    "start": 464.23,
                    "end": 464.49,
                    "word": " it",
                    "probability": 0.900390625
                },
                {
                    "start": 464.49,
                    "end": 464.65,
                    "word": " is",
                    "probability": 0.8955078125
                },
                {
                    "start": 464.65,
                    "end": 464.91,
                    "word": " less",
                    "probability": 0.88671875
                },
                {
                    "start": 464.91,
                    "end": 465.41,
                    "word": " precise",
                    "probability": 0.89794921875
                },
                {
                    "start": 465.41,
                    "end": 466.03,
                    "word": ",",
                    "probability": 0.7880859375
                },
                {
                    "start": 466.03,
                    "end": 466.17,
                    "word": " but",
                    "probability": 0.89697265625
                },
                {
                    "start": 466.17,
                    "end": 466.29,
                    "word": " you",
                    "probability": 0.89501953125
                },
                {
                    "start": 466.29,
                    "end": 466.53,
                    "word": " still",
                    "probability": 0.89501953125
                },
                {
                    "start": 466.53,
                    "end": 466.87,
                    "word": " get",
                    "probability": 0.8837890625
                },
                {
                    "start": 466.87,
                    "end": 467.05,
                    "word": " the",
                    "probability": 0.80224609375
                },
                {
                    "start": 467.05,
                    "end": 467.25,
                    "word": " same",
                    "probability": 0.84912109375
                },
                {
                    "start": 467.25,
                    "end": 467.73,
                    "word": " insights",
                    "probability": 0.81396484375
                },
                {
                    "start": 467.73,
                    "end": 468.01,
                    "word": " about",
                    "probability": 0.80224609375
                },
                {
                    "start": 468.01,
                    "end": 468.21,
                    "word": " the",
                    "probability": 0.80859375
                },
                {
                    "start": 468.21,
                    "end": 468.65,
                    "word": " statistical",
                    "probability": 0.8486328125
                },
                {
                    "start": 468.65,
                    "end": 469.27,
                    "word": " trends",
                    "probability": 0.7109375
                },
                {
                    "start": 469.27,
                    "end": 469.63,
                    "word": ".",
                    "probability": 0.953125
                }
            ]
        },
        {
            "id": 62,
            "seek": 48547,
            "start": 472.17,
            "end": 487.51,
            "text": " Similarly, let's look at day two. Same thing. But now, if you compare the stats for day one and day two, you can't tell which country the single person is coming from.",
            "tokens": [
                50364,
                13157,
                11,
                718,
                311,
                574,
                412,
                786,
                732,
                13,
                10635,
                551,
                13,
                583,
                586,
                11,
                498,
                291,
                6794,
                264,
                18152,
                337,
                786,
                472,
                293,
                786,
                732,
                11,
                291,
                393,
                380,
                980,
                597,
                1941,
                264,
                2167,
                954,
                307,
                1348,
                490,
                13,
                51214
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20632102645256303,
            "compression_ratio": 1.336,
            "no_speech_prob": 0.003765106201171875,
            "words": [
                {
                    "start": 472.17,
                    "end": 472.77,
                    "word": " Similarly",
                    "probability": 0.7216796875
                },
                {
                    "start": 472.77,
                    "end": 473.17,
                    "word": ",",
                    "probability": 0.884765625
                },
                {
                    "start": 473.17,
                    "end": 473.77,
                    "word": " let's",
                    "probability": 0.942626953125
                },
                {
                    "start": 473.77,
                    "end": 473.95,
                    "word": " look",
                    "probability": 0.9208984375
                },
                {
                    "start": 473.95,
                    "end": 474.11,
                    "word": " at",
                    "probability": 0.90380859375
                },
                {
                    "start": 474.11,
                    "end": 474.29,
                    "word": " day",
                    "probability": 0.74072265625
                },
                {
                    "start": 474.29,
                    "end": 474.57,
                    "word": " two",
                    "probability": 0.564453125
                },
                {
                    "start": 474.57,
                    "end": 475.17,
                    "word": ".",
                    "probability": 0.69775390625
                },
                {
                    "start": 475.97,
                    "end": 476.43,
                    "word": " Same",
                    "probability": 0.81005859375
                },
                {
                    "start": 476.43,
                    "end": 476.79,
                    "word": " thing",
                    "probability": 0.81396484375
                },
                {
                    "start": 476.79,
                    "end": 477.39,
                    "word": ".",
                    "probability": 0.75048828125
                },
                {
                    "start": 478.35,
                    "end": 478.57,
                    "word": " But",
                    "probability": 0.828125
                },
                {
                    "start": 478.57,
                    "end": 478.85,
                    "word": " now",
                    "probability": 0.87841796875
                },
                {
                    "start": 478.85,
                    "end": 479.05,
                    "word": ",",
                    "probability": 0.6494140625
                },
                {
                    "start": 479.05,
                    "end": 479.33,
                    "word": " if",
                    "probability": 0.90234375
                },
                {
                    "start": 479.33,
                    "end": 479.45,
                    "word": " you",
                    "probability": 0.89208984375
                },
                {
                    "start": 479.45,
                    "end": 479.77,
                    "word": " compare",
                    "probability": 0.85546875
                },
                {
                    "start": 479.77,
                    "end": 479.97,
                    "word": " the",
                    "probability": 0.80517578125
                },
                {
                    "start": 479.97,
                    "end": 480.27,
                    "word": " stats",
                    "probability": 0.83154296875
                },
                {
                    "start": 480.27,
                    "end": 480.53,
                    "word": " for",
                    "probability": 0.896484375
                },
                {
                    "start": 480.53,
                    "end": 480.77,
                    "word": " day",
                    "probability": 0.8759765625
                },
                {
                    "start": 480.77,
                    "end": 481.07,
                    "word": " one",
                    "probability": 0.7978515625
                },
                {
                    "start": 481.07,
                    "end": 481.47,
                    "word": " and",
                    "probability": 0.8857421875
                },
                {
                    "start": 481.47,
                    "end": 481.71,
                    "word": " day",
                    "probability": 0.8828125
                },
                {
                    "start": 481.71,
                    "end": 482.01,
                    "word": " two",
                    "probability": 0.8232421875
                },
                {
                    "start": 482.01,
                    "end": 482.21,
                    "word": ",",
                    "probability": 0.9443359375
                },
                {
                    "start": 482.21,
                    "end": 482.59,
                    "word": " you",
                    "probability": 0.89453125
                },
                {
                    "start": 482.59,
                    "end": 483.17,
                    "word": " can't",
                    "probability": 0.936767578125
                },
                {
                    "start": 483.17,
                    "end": 483.41,
                    "word": " tell",
                    "probability": 0.80859375
                },
                {
                    "start": 483.41,
                    "end": 483.71,
                    "word": " which",
                    "probability": 0.90380859375
                },
                {
                    "start": 483.71,
                    "end": 484.21,
                    "word": " country",
                    "probability": 0.80859375
                },
                {
                    "start": 484.21,
                    "end": 484.91,
                    "word": " the",
                    "probability": 0.62548828125
                },
                {
                    "start": 484.91,
                    "end": 485.23,
                    "word": " single",
                    "probability": 0.87646484375
                },
                {
                    "start": 485.23,
                    "end": 485.73,
                    "word": " person",
                    "probability": 0.80859375
                },
                {
                    "start": 485.73,
                    "end": 486.49,
                    "word": " is",
                    "probability": 0.8974609375
                },
                {
                    "start": 486.49,
                    "end": 486.75,
                    "word": " coming",
                    "probability": 0.80712890625
                },
                {
                    "start": 486.75,
                    "end": 487.37,
                    "word": " from",
                    "probability": 0.8173828125
                },
                {
                    "start": 487.37,
                    "end": 487.51,
                    "word": ".",
                    "probability": 0.70703125
                }
            ]
        },
        {
            "id": 63,
            "seek": 51247,
            "start": 487.51,
            "end": 495.41,
            "text": " And you know it's not worth trying since you know the numbers were changed a little. This is the principle behind differential privacy.",
            "tokens": [
                50364,
                400,
                291,
                458,
                309,
                311,
                406,
                3163,
                1382,
                1670,
                291,
                458,
                264,
                3547,
                645,
                3105,
                257,
                707,
                13,
                639,
                307,
                264,
                8665,
                2261,
                15756,
                11427,
                13,
                50714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2447265662252903,
            "compression_ratio": 1.6465863453815262,
            "no_speech_prob": 0.3310546875,
            "words": [
                {
                    "start": 487.51,
                    "end": 487.59,
                    "word": " And",
                    "probability": 0.4375
                },
                {
                    "start": 487.59,
                    "end": 487.71,
                    "word": " you",
                    "probability": 0.56689453125
                },
                {
                    "start": 487.71,
                    "end": 487.83,
                    "word": " know",
                    "probability": 0.79931640625
                },
                {
                    "start": 487.83,
                    "end": 487.95,
                    "word": " it's",
                    "probability": 0.87158203125
                },
                {
                    "start": 487.95,
                    "end": 488.19,
                    "word": " not",
                    "probability": 0.876953125
                },
                {
                    "start": 488.19,
                    "end": 488.57,
                    "word": " worth",
                    "probability": 0.89306640625
                },
                {
                    "start": 488.57,
                    "end": 488.97,
                    "word": " trying",
                    "probability": 0.76611328125
                },
                {
                    "start": 488.97,
                    "end": 489.33,
                    "word": " since",
                    "probability": 0.59765625
                },
                {
                    "start": 489.33,
                    "end": 489.53,
                    "word": " you",
                    "probability": 0.86474609375
                },
                {
                    "start": 489.53,
                    "end": 489.69,
                    "word": " know",
                    "probability": 0.83203125
                },
                {
                    "start": 489.69,
                    "end": 489.83,
                    "word": " the",
                    "probability": 0.76953125
                },
                {
                    "start": 489.83,
                    "end": 490.15,
                    "word": " numbers",
                    "probability": 0.8544921875
                },
                {
                    "start": 490.15,
                    "end": 490.37,
                    "word": " were",
                    "probability": 0.572265625
                },
                {
                    "start": 490.37,
                    "end": 490.83,
                    "word": " changed",
                    "probability": 0.80126953125
                },
                {
                    "start": 491.11,
                    "end": 491.31,
                    "word": " a little",
                    "probability": 0.80419921875
                },
                {
                    "start": 491.31,
                    "end": 491.99,
                    "word": ".",
                    "probability": 0.93310546875
                },
                {
                    "start": 492.31,
                    "end": 492.59,
                    "word": " This",
                    "probability": 0.8134765625
                },
                {
                    "start": 492.59,
                    "end": 492.71,
                    "word": " is",
                    "probability": 0.8984375
                },
                {
                    "start": 492.71,
                    "end": 492.87,
                    "word": " the",
                    "probability": 0.806640625
                },
                {
                    "start": 492.87,
                    "end": 493.23,
                    "word": " principle",
                    "probability": 0.78125
                },
                {
                    "start": 493.23,
                    "end": 493.81,
                    "word": " behind",
                    "probability": 0.85400390625
                },
                {
                    "start": 493.81,
                    "end": 494.27,
                    "word": " differential",
                    "probability": 0.69775390625
                },
                {
                    "start": 494.27,
                    "end": 494.73,
                    "word": " privacy",
                    "probability": 0.64990234375
                },
                {
                    "start": 494.73,
                    "end": 495.41,
                    "word": ".",
                    "probability": 0.935546875
                }
            ]
        },
        {
            "id": 64,
            "seek": 51247,
            "start": 495.57,
            "end": 514.51,
            "text": " Differential privacy introduces the right amount of noise to the data, so we're able to provide useful statistically significant insights while ensuring that no one can tell if a particular individual's information was used in the competition in a mathematically proven way.",
            "tokens": [
                50764,
                413,
                12612,
                2549,
                11427,
                31472,
                264,
                558,
                2372,
                295,
                5658,
                281,
                264,
                1412,
                11,
                370,
                321,
                434,
                1075,
                281,
                2893,
                4420,
                36478,
                4776,
                14310,
                1339,
                16882,
                300,
                572,
                472,
                393,
                980,
                498,
                257,
                1729,
                2609,
                311,
                1589,
                390,
                1143,
                294,
                264,
                6211,
                294,
                257,
                44003,
                12785,
                636,
                13,
                51664
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2447265662252903,
            "compression_ratio": 1.6465863453815262,
            "no_speech_prob": 0.3310546875,
            "words": [
                {
                    "start": 495.57,
                    "end": 495.95,
                    "word": " Differential",
                    "probability": 0.9202473958333334
                },
                {
                    "start": 495.95,
                    "end": 496.43,
                    "word": " privacy",
                    "probability": 0.79541015625
                },
                {
                    "start": 496.43,
                    "end": 497.17,
                    "word": " introduces",
                    "probability": 0.7509765625
                },
                {
                    "start": 497.17,
                    "end": 497.45,
                    "word": " the",
                    "probability": 0.80810546875
                },
                {
                    "start": 497.45,
                    "end": 497.77,
                    "word": " right",
                    "probability": 0.8837890625
                },
                {
                    "start": 497.77,
                    "end": 498.13,
                    "word": " amount",
                    "probability": 0.8359375
                },
                {
                    "start": 498.13,
                    "end": 498.35,
                    "word": " of",
                    "probability": 0.9033203125
                },
                {
                    "start": 498.35,
                    "end": 498.67,
                    "word": " noise",
                    "probability": 0.904296875
                },
                {
                    "start": 498.67,
                    "end": 498.89,
                    "word": " to",
                    "probability": 0.90087890625
                },
                {
                    "start": 498.89,
                    "end": 499.07,
                    "word": " the",
                    "probability": 0.80908203125
                },
                {
                    "start": 499.07,
                    "end": 499.43,
                    "word": " data",
                    "probability": 0.82373046875
                },
                {
                    "start": 499.43,
                    "end": 500.05,
                    "word": ",",
                    "probability": 0.3505859375
                },
                {
                    "start": 500.05,
                    "end": 500.15,
                    "word": " so",
                    "probability": 0.87646484375
                },
                {
                    "start": 500.15,
                    "end": 501.05,
                    "word": " we're",
                    "probability": 0.8037109375
                },
                {
                    "start": 501.05,
                    "end": 501.31,
                    "word": " able",
                    "probability": 0.8701171875
                },
                {
                    "start": 501.31,
                    "end": 501.53,
                    "word": " to",
                    "probability": 0.90869140625
                },
                {
                    "start": 501.53,
                    "end": 501.89,
                    "word": " provide",
                    "probability": 0.90869140625
                },
                {
                    "start": 501.89,
                    "end": 502.63,
                    "word": " useful",
                    "probability": 0.7412109375
                },
                {
                    "start": 502.63,
                    "end": 503.37,
                    "word": " statistically",
                    "probability": 0.34619140625
                },
                {
                    "start": 503.37,
                    "end": 504.19,
                    "word": " significant",
                    "probability": 0.7822265625
                },
                {
                    "start": 504.19,
                    "end": 504.75,
                    "word": " insights",
                    "probability": 0.8759765625
                },
                {
                    "start": 504.75,
                    "end": 505.37,
                    "word": " while",
                    "probability": 0.7265625
                },
                {
                    "start": 505.37,
                    "end": 505.73,
                    "word": " ensuring",
                    "probability": 0.83251953125
                },
                {
                    "start": 505.73,
                    "end": 505.97,
                    "word": " that",
                    "probability": 0.87060546875
                },
                {
                    "start": 505.97,
                    "end": 506.27,
                    "word": " no",
                    "probability": 0.9072265625
                },
                {
                    "start": 506.27,
                    "end": 506.47,
                    "word": " one",
                    "probability": 0.81298828125
                },
                {
                    "start": 506.47,
                    "end": 506.73,
                    "word": " can",
                    "probability": 0.90185546875
                },
                {
                    "start": 506.73,
                    "end": 507.03,
                    "word": " tell",
                    "probability": 0.8212890625
                },
                {
                    "start": 507.03,
                    "end": 507.27,
                    "word": " if",
                    "probability": 0.892578125
                },
                {
                    "start": 507.43,
                    "end": 507.83,
                    "word": " a particular",
                    "probability": 0.810546875
                },
                {
                    "start": 507.83,
                    "end": 508.59,
                    "word": " individual's",
                    "probability": 0.831298828125
                },
                {
                    "start": 508.59,
                    "end": 509.11,
                    "word": " information",
                    "probability": 0.73828125
                },
                {
                    "start": 509.11,
                    "end": 509.91,
                    "word": " was",
                    "probability": 0.89453125
                },
                {
                    "start": 509.91,
                    "end": 510.25,
                    "word": " used",
                    "probability": 0.8251953125
                },
                {
                    "start": 510.25,
                    "end": 510.57,
                    "word": " in",
                    "probability": 0.90478515625
                },
                {
                    "start": 510.57,
                    "end": 510.83,
                    "word": " the",
                    "probability": 0.8037109375
                },
                {
                    "start": 510.83,
                    "end": 511.39,
                    "word": " competition",
                    "probability": 0.2099609375
                },
                {
                    "start": 511.39,
                    "end": 512.07,
                    "word": " in",
                    "probability": 0.73046875
                },
                {
                    "start": 512.21,
                    "end": 512.67,
                    "word": " a mathematically",
                    "probability": 0.802734375
                },
                {
                    "start": 512.67,
                    "end": 513.27,
                    "word": " proven",
                    "probability": 0.8310546875
                },
                {
                    "start": 513.27,
                    "end": 513.83,
                    "word": " way",
                    "probability": 0.8876953125
                },
                {
                    "start": 513.83,
                    "end": 514.51,
                    "word": ".",
                    "probability": 0.994140625
                }
            ]
        },
        {
            "id": 65,
            "seek": 53995,
            "start": 515.81,
            "end": 527.95,
            "text": " Just to recap, and Brian will talk a little bit more about it in a bit, an algorithm is differentially private if its output doesn't change much when a single person is added to the database.",
            "tokens": [
                50414,
                1449,
                281,
                20928,
                11,
                293,
                10765,
                486,
                751,
                257,
                707,
                857,
                544,
                466,
                309,
                294,
                257,
                857,
                11,
                364,
                9284,
                307,
                15756,
                356,
                4551,
                498,
                1080,
                5598,
                1177,
                380,
                1319,
                709,
                562,
                257,
                2167,
                954,
                307,
                3869,
                281,
                264,
                8149,
                13,
                50964
            ],
            "temperature": 0.0,
            "avg_logprob": -0.23139881608741625,
            "compression_ratio": 1.612,
            "no_speech_prob": 0.206787109375,
            "words": [
                {
                    "start": 515.81,
                    "end": 516.45,
                    "word": " Just",
                    "probability": 0.85498046875
                },
                {
                    "start": 516.45,
                    "end": 516.65,
                    "word": " to",
                    "probability": 0.8671875
                },
                {
                    "start": 516.65,
                    "end": 516.99,
                    "word": " recap",
                    "probability": 0.755859375
                },
                {
                    "start": 516.99,
                    "end": 517.29,
                    "word": ",",
                    "probability": 0.611328125
                },
                {
                    "start": 517.29,
                    "end": 517.31,
                    "word": " and",
                    "probability": 0.83740234375
                },
                {
                    "start": 517.31,
                    "end": 517.67,
                    "word": " Brian",
                    "probability": 0.7080078125
                },
                {
                    "start": 517.67,
                    "end": 517.89,
                    "word": " will",
                    "probability": 0.79443359375
                },
                {
                    "start": 517.89,
                    "end": 518.17,
                    "word": " talk",
                    "probability": 0.80712890625
                },
                {
                    "start": 518.31,
                    "end": 518.43,
                    "word": " a little",
                    "probability": 0.798828125
                },
                {
                    "start": 518.43,
                    "end": 518.63,
                    "word": " bit",
                    "probability": 0.8701171875
                },
                {
                    "start": 518.63,
                    "end": 518.85,
                    "word": " more",
                    "probability": 0.87744140625
                },
                {
                    "start": 518.85,
                    "end": 519.17,
                    "word": " about",
                    "probability": 0.8134765625
                },
                {
                    "start": 519.17,
                    "end": 519.47,
                    "word": " it",
                    "probability": 0.89892578125
                },
                {
                    "start": 519.47,
                    "end": 519.79,
                    "word": " in",
                    "probability": 0.81201171875
                },
                {
                    "start": 519.89,
                    "end": 520.13,
                    "word": " a bit",
                    "probability": 0.90478515625
                },
                {
                    "start": 520.13,
                    "end": 520.59,
                    "word": ",",
                    "probability": 0.6982421875
                },
                {
                    "start": 520.59,
                    "end": 520.79,
                    "word": " an",
                    "probability": 0.85546875
                },
                {
                    "start": 520.79,
                    "end": 521.27,
                    "word": " algorithm",
                    "probability": 0.8173828125
                },
                {
                    "start": 521.27,
                    "end": 521.63,
                    "word": " is",
                    "probability": 0.89892578125
                },
                {
                    "start": 521.63,
                    "end": 522.23,
                    "word": " differentially",
                    "probability": 0.5872802734375
                },
                {
                    "start": 522.23,
                    "end": 522.51,
                    "word": " private",
                    "probability": 0.48583984375
                },
                {
                    "start": 522.51,
                    "end": 522.83,
                    "word": " if",
                    "probability": 0.88037109375
                },
                {
                    "start": 522.83,
                    "end": 523.03,
                    "word": " its",
                    "probability": 0.77490234375
                },
                {
                    "start": 523.03,
                    "end": 523.37,
                    "word": " output",
                    "probability": 0.8330078125
                },
                {
                    "start": 523.37,
                    "end": 523.77,
                    "word": " doesn't",
                    "probability": 0.92822265625
                },
                {
                    "start": 523.77,
                    "end": 524.19,
                    "word": " change",
                    "probability": 0.8427734375
                },
                {
                    "start": 524.19,
                    "end": 524.63,
                    "word": " much",
                    "probability": 0.80908203125
                },
                {
                    "start": 524.63,
                    "end": 525.31,
                    "word": " when",
                    "probability": 0.76123046875
                },
                {
                    "start": 525.49,
                    "end": 525.73,
                    "word": " a single",
                    "probability": 0.87841796875
                },
                {
                    "start": 525.73,
                    "end": 526.07,
                    "word": " person",
                    "probability": 0.78857421875
                },
                {
                    "start": 526.07,
                    "end": 526.29,
                    "word": " is",
                    "probability": 0.90771484375
                },
                {
                    "start": 526.29,
                    "end": 526.53,
                    "word": " added",
                    "probability": 0.80078125
                },
                {
                    "start": 526.53,
                    "end": 526.73,
                    "word": " to",
                    "probability": 0.908203125
                },
                {
                    "start": 526.73,
                    "end": 526.87,
                    "word": " the",
                    "probability": 0.611328125
                },
                {
                    "start": 526.87,
                    "end": 527.31,
                    "word": " database",
                    "probability": 0.7734375
                },
                {
                    "start": 527.31,
                    "end": 527.95,
                    "word": ".",
                    "probability": 0.9345703125
                }
            ]
        },
        {
            "id": 66,
            "seek": 53995,
            "start": 528.01,
            "end": 541.99,
            "text": " This has a rigorous mathematical definition, which basically considers a pair of two neighboring data sets, calculates the output of those data sets, and defines a threshold of the probability density functions.",
            "tokens": [
                51014,
                639,
                575,
                257,
                29882,
                18894,
                7123,
                11,
                597,
                1936,
                33095,
                257,
                6119,
                295,
                732,
                31521,
                1412,
                6352,
                11,
                4322,
                1024,
                264,
                5598,
                295,
                729,
                1412,
                6352,
                11,
                293,
                23122,
                257,
                14678,
                295,
                264,
                8482,
                10305,
                6828,
                13,
                51664
            ],
            "temperature": 0.0,
            "avg_logprob": -0.23139881608741625,
            "compression_ratio": 1.612,
            "no_speech_prob": 0.206787109375,
            "words": [
                {
                    "start": 528.01,
                    "end": 528.25,
                    "word": " This",
                    "probability": 0.78125
                },
                {
                    "start": 528.25,
                    "end": 528.57,
                    "word": " has",
                    "probability": 0.900390625
                },
                {
                    "start": 528.77,
                    "end": 529.11,
                    "word": " a rigorous",
                    "probability": 0.861328125
                },
                {
                    "start": 529.11,
                    "end": 529.81,
                    "word": " mathematical",
                    "probability": 0.73828125
                },
                {
                    "start": 529.81,
                    "end": 530.35,
                    "word": " definition",
                    "probability": 0.828125
                },
                {
                    "start": 530.35,
                    "end": 530.61,
                    "word": ",",
                    "probability": 0.368408203125
                },
                {
                    "start": 530.61,
                    "end": 530.73,
                    "word": " which",
                    "probability": 0.91845703125
                },
                {
                    "start": 530.73,
                    "end": 531.27,
                    "word": " basically",
                    "probability": 0.8974609375
                },
                {
                    "start": 531.27,
                    "end": 532.21,
                    "word": " considers",
                    "probability": 0.75634765625
                },
                {
                    "start": 532.51,
                    "end": 532.79,
                    "word": " a pair",
                    "probability": 0.84521484375
                },
                {
                    "start": 532.79,
                    "end": 533.11,
                    "word": " of",
                    "probability": 0.90869140625
                },
                {
                    "start": 533.11,
                    "end": 533.31,
                    "word": " two",
                    "probability": 0.791015625
                },
                {
                    "start": 533.31,
                    "end": 533.83,
                    "word": " neighboring",
                    "probability": 0.7265625
                },
                {
                    "start": 533.83,
                    "end": 534.11,
                    "word": " data",
                    "probability": 0.34619140625
                },
                {
                    "start": 534.11,
                    "end": 534.47,
                    "word": " sets",
                    "probability": 0.7978515625
                },
                {
                    "start": 534.47,
                    "end": 535.07,
                    "word": ",",
                    "probability": 0.439208984375
                },
                {
                    "start": 535.07,
                    "end": 535.81,
                    "word": " calculates",
                    "probability": 0.73828125
                },
                {
                    "start": 535.81,
                    "end": 535.99,
                    "word": " the",
                    "probability": 0.81298828125
                },
                {
                    "start": 535.99,
                    "end": 536.29,
                    "word": " output",
                    "probability": 0.83056640625
                },
                {
                    "start": 536.29,
                    "end": 536.49,
                    "word": " of",
                    "probability": 0.908203125
                },
                {
                    "start": 536.49,
                    "end": 536.79,
                    "word": " those",
                    "probability": 0.84765625
                },
                {
                    "start": 536.79,
                    "end": 537.31,
                    "word": " data",
                    "probability": 0.71142578125
                },
                {
                    "start": 537.31,
                    "end": 537.67,
                    "word": " sets",
                    "probability": 0.78759765625
                },
                {
                    "start": 537.67,
                    "end": 537.99,
                    "word": ",",
                    "probability": 0.71630859375
                },
                {
                    "start": 537.99,
                    "end": 538.31,
                    "word": " and",
                    "probability": 0.8935546875
                },
                {
                    "start": 538.31,
                    "end": 538.63,
                    "word": " defines",
                    "probability": 0.61083984375
                },
                {
                    "start": 538.83,
                    "end": 539.17,
                    "word": " a threshold",
                    "probability": 0.5947265625
                },
                {
                    "start": 539.17,
                    "end": 539.51,
                    "word": " of",
                    "probability": 0.89892578125
                },
                {
                    "start": 539.51,
                    "end": 539.67,
                    "word": " the",
                    "probability": 0.79833984375
                },
                {
                    "start": 539.67,
                    "end": 540.25,
                    "word": " probability",
                    "probability": 0.88623046875
                },
                {
                    "start": 540.25,
                    "end": 540.71,
                    "word": " density",
                    "probability": 0.83447265625
                },
                {
                    "start": 540.71,
                    "end": 541.35,
                    "word": " functions",
                    "probability": 0.7666015625
                },
                {
                    "start": 541.35,
                    "end": 541.99,
                    "word": ".",
                    "probability": 0.984375
                }
            ]
        },
        {
            "id": 67,
            "seek": 56785,
            "start": 541.99,
            "end": 548.45,
            "text": " This threshold is what defines how much the output is allowed to change, and it's called epsilon.",
            "tokens": [
                50364,
                639,
                14678,
                307,
                437,
                23122,
                577,
                709,
                264,
                5598,
                307,
                4350,
                281,
                1319,
                11,
                293,
                309,
                311,
                1219,
                17889,
                13,
                50664
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19593750491738318,
            "compression_ratio": 1.5671641791044777,
            "no_speech_prob": 0.029296875,
            "words": [
                {
                    "start": 541.99,
                    "end": 542.27,
                    "word": " This",
                    "probability": 0.6708984375
                },
                {
                    "start": 542.27,
                    "end": 542.83,
                    "word": " threshold",
                    "probability": 0.59423828125
                },
                {
                    "start": 542.83,
                    "end": 543.37,
                    "word": " is",
                    "probability": 0.5693359375
                },
                {
                    "start": 543.37,
                    "end": 543.71,
                    "word": " what",
                    "probability": 0.859375
                },
                {
                    "start": 543.71,
                    "end": 544.07,
                    "word": " defines",
                    "probability": 0.7783203125
                },
                {
                    "start": 544.07,
                    "end": 544.39,
                    "word": " how",
                    "probability": 0.896484375
                },
                {
                    "start": 544.39,
                    "end": 544.69,
                    "word": " much",
                    "probability": 0.79833984375
                },
                {
                    "start": 544.69,
                    "end": 544.87,
                    "word": " the",
                    "probability": 0.7998046875
                },
                {
                    "start": 544.87,
                    "end": 545.13,
                    "word": " output",
                    "probability": 0.7958984375
                },
                {
                    "start": 545.13,
                    "end": 545.39,
                    "word": " is",
                    "probability": 0.90576171875
                },
                {
                    "start": 545.39,
                    "end": 545.69,
                    "word": " allowed",
                    "probability": 0.84765625
                },
                {
                    "start": 545.69,
                    "end": 545.87,
                    "word": " to",
                    "probability": 0.9052734375
                },
                {
                    "start": 545.87,
                    "end": 546.29,
                    "word": " change",
                    "probability": 0.83447265625
                },
                {
                    "start": 546.29,
                    "end": 546.75,
                    "word": ",",
                    "probability": 0.394775390625
                },
                {
                    "start": 546.75,
                    "end": 546.85,
                    "word": " and",
                    "probability": 0.8564453125
                },
                {
                    "start": 546.85,
                    "end": 547.05,
                    "word": " it's",
                    "probability": 0.824951171875
                },
                {
                    "start": 547.05,
                    "end": 547.31,
                    "word": " called",
                    "probability": 0.81494140625
                },
                {
                    "start": 547.31,
                    "end": 547.85,
                    "word": " epsilon",
                    "probability": 0.465576171875
                },
                {
                    "start": 547.85,
                    "end": 548.45,
                    "word": ".",
                    "probability": 0.96240234375
                }
            ]
        },
        {
            "id": 68,
            "seek": 56785,
            "start": 549.05,
            "end": 557.01,
            "text": " Now, let me talk to you about some open sourcing efforts we've done at Google to share some of this technology with the world.",
            "tokens": [
                50714,
                823,
                11,
                718,
                385,
                751,
                281,
                291,
                466,
                512,
                1269,
                11006,
                2175,
                6484,
                321,
                600,
                1096,
                412,
                3329,
                281,
                2073,
                512,
                295,
                341,
                2899,
                365,
                264,
                1002,
                13,
                51064
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19593750491738318,
            "compression_ratio": 1.5671641791044777,
            "no_speech_prob": 0.029296875,
            "words": [
                {
                    "start": 549.05,
                    "end": 549.65,
                    "word": " Now",
                    "probability": 0.85986328125
                },
                {
                    "start": 549.65,
                    "end": 550.05,
                    "word": ",",
                    "probability": 0.67724609375
                },
                {
                    "start": 550.05,
                    "end": 550.29,
                    "word": " let",
                    "probability": 0.9072265625
                },
                {
                    "start": 550.29,
                    "end": 550.39,
                    "word": " me",
                    "probability": 0.90087890625
                },
                {
                    "start": 550.39,
                    "end": 550.61,
                    "word": " talk",
                    "probability": 0.80859375
                },
                {
                    "start": 550.61,
                    "end": 550.73,
                    "word": " to",
                    "probability": 0.90283203125
                },
                {
                    "start": 550.73,
                    "end": 550.95,
                    "word": " you",
                    "probability": 0.8935546875
                },
                {
                    "start": 550.95,
                    "end": 551.25,
                    "word": " about",
                    "probability": 0.82421875
                },
                {
                    "start": 551.25,
                    "end": 551.65,
                    "word": " some",
                    "probability": 0.81005859375
                },
                {
                    "start": 551.65,
                    "end": 552.13,
                    "word": " open",
                    "probability": 0.748046875
                },
                {
                    "start": 552.13,
                    "end": 552.45,
                    "word": " sourcing",
                    "probability": 0.85888671875
                },
                {
                    "start": 552.45,
                    "end": 552.89,
                    "word": " efforts",
                    "probability": 0.8125
                },
                {
                    "start": 552.89,
                    "end": 553.17,
                    "word": " we've",
                    "probability": 0.8876953125
                },
                {
                    "start": 553.17,
                    "end": 553.35,
                    "word": " done",
                    "probability": 0.7890625
                },
                {
                    "start": 553.35,
                    "end": 553.57,
                    "word": " at",
                    "probability": 0.87158203125
                },
                {
                    "start": 553.57,
                    "end": 553.87,
                    "word": " Google",
                    "probability": 0.88037109375
                },
                {
                    "start": 553.87,
                    "end": 554.17,
                    "word": " to",
                    "probability": 0.880859375
                },
                {
                    "start": 554.17,
                    "end": 554.49,
                    "word": " share",
                    "probability": 0.8125
                },
                {
                    "start": 554.49,
                    "end": 554.85,
                    "word": " some",
                    "probability": 0.798828125
                },
                {
                    "start": 554.85,
                    "end": 555.01,
                    "word": " of",
                    "probability": 0.9072265625
                },
                {
                    "start": 555.01,
                    "end": 555.15,
                    "word": " this",
                    "probability": 0.84375
                },
                {
                    "start": 555.15,
                    "end": 555.67,
                    "word": " technology",
                    "probability": 0.9140625
                },
                {
                    "start": 555.67,
                    "end": 556.09,
                    "word": " with",
                    "probability": 0.806640625
                },
                {
                    "start": 556.09,
                    "end": 556.29,
                    "word": " the",
                    "probability": 0.81640625
                },
                {
                    "start": 556.29,
                    "end": 556.57,
                    "word": " world",
                    "probability": 0.88525390625
                },
                {
                    "start": 556.57,
                    "end": 557.01,
                    "word": ".",
                    "probability": 0.98876953125
                }
            ]
        },
        {
            "id": 69,
            "seek": 56785,
            "start": 557.01,
            "end": 562.75,
            "text": " We publish our code in an open source library so that others can reuse our work.",
            "tokens": [
                51114,
                492,
                11374,
                527,
                3089,
                294,
                364,
                1269,
                4009,
                6405,
                370,
                300,
                2357,
                393,
                26225,
                527,
                589,
                13,
                51364
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19593750491738318,
            "compression_ratio": 1.5671641791044777,
            "no_speech_prob": 0.029296875,
            "words": [
                {
                    "start": 557.01,
                    "end": 557.43,
                    "word": " We",
                    "probability": 0.9013671875
                },
                {
                    "start": 557.43,
                    "end": 557.77,
                    "word": " publish",
                    "probability": 0.8154296875
                },
                {
                    "start": 557.77,
                    "end": 558.07,
                    "word": " our",
                    "probability": 0.8154296875
                },
                {
                    "start": 558.07,
                    "end": 558.53,
                    "word": " code",
                    "probability": 0.81494140625
                },
                {
                    "start": 558.53,
                    "end": 559.01,
                    "word": " in",
                    "probability": 0.90185546875
                },
                {
                    "start": 559.01,
                    "end": 559.21,
                    "word": " an",
                    "probability": 0.861328125
                },
                {
                    "start": 559.21,
                    "end": 559.51,
                    "word": " open",
                    "probability": 0.794921875
                },
                {
                    "start": 559.51,
                    "end": 559.73,
                    "word": " source",
                    "probability": 0.61572265625
                },
                {
                    "start": 559.73,
                    "end": 560.23,
                    "word": " library",
                    "probability": 0.80224609375
                },
                {
                    "start": 560.23,
                    "end": 560.57,
                    "word": " so",
                    "probability": 0.81005859375
                },
                {
                    "start": 560.57,
                    "end": 560.75,
                    "word": " that",
                    "probability": 0.87060546875
                },
                {
                    "start": 560.75,
                    "end": 561.01,
                    "word": " others",
                    "probability": 0.7509765625
                },
                {
                    "start": 561.01,
                    "end": 561.23,
                    "word": " can",
                    "probability": 0.90087890625
                },
                {
                    "start": 561.23,
                    "end": 561.57,
                    "word": " reuse",
                    "probability": 0.73095703125
                },
                {
                    "start": 561.57,
                    "end": 561.83,
                    "word": " our",
                    "probability": 0.82421875
                },
                {
                    "start": 561.83,
                    "end": 562.17,
                    "word": " work",
                    "probability": 0.8134765625
                },
                {
                    "start": 562.17,
                    "end": 562.75,
                    "word": ".",
                    "probability": 0.99169921875
                }
            ]
        },
        {
            "id": 70,
            "seek": 56785,
            "start": 562.75,
            "end": 569.89,
            "text": " As I mentioned before, differential privacy has a lot of implementation subtopics, and it is hard to get it right.",
            "tokens": [
                51414,
                1018,
                286,
                2835,
                949,
                11,
                15756,
                11427,
                575,
                257,
                688,
                295,
                11420,
                7257,
                404,
                1167,
                11,
                293,
                309,
                307,
                1152,
                281,
                483,
                309,
                558,
                13,
                51714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19593750491738318,
            "compression_ratio": 1.5671641791044777,
            "no_speech_prob": 0.029296875,
            "words": [
                {
                    "start": 562.75,
                    "end": 563.15,
                    "word": " As",
                    "probability": 0.8935546875
                },
                {
                    "start": 563.15,
                    "end": 563.27,
                    "word": " I",
                    "probability": 0.99560546875
                },
                {
                    "start": 563.27,
                    "end": 563.55,
                    "word": " mentioned",
                    "probability": 0.763671875
                },
                {
                    "start": 563.55,
                    "end": 564.01,
                    "word": " before",
                    "probability": 0.78759765625
                },
                {
                    "start": 564.01,
                    "end": 564.33,
                    "word": ",",
                    "probability": 0.98193359375
                },
                {
                    "start": 564.33,
                    "end": 564.61,
                    "word": " differential",
                    "probability": 0.62255859375
                },
                {
                    "start": 564.61,
                    "end": 565.03,
                    "word": " privacy",
                    "probability": 0.80029296875
                },
                {
                    "start": 565.03,
                    "end": 565.35,
                    "word": " has",
                    "probability": 0.89599609375
                },
                {
                    "start": 565.51,
                    "end": 565.65,
                    "word": " a lot",
                    "probability": 0.8935546875
                },
                {
                    "start": 565.65,
                    "end": 565.81,
                    "word": " of",
                    "probability": 0.89697265625
                },
                {
                    "start": 565.81,
                    "end": 566.69,
                    "word": " implementation",
                    "probability": 0.59765625
                },
                {
                    "start": 566.69,
                    "end": 567.45,
                    "word": " subtopics",
                    "probability": 0.6795247395833334
                },
                {
                    "start": 567.45,
                    "end": 567.71,
                    "word": ",",
                    "probability": 0.64599609375
                },
                {
                    "start": 567.71,
                    "end": 568.03,
                    "word": " and",
                    "probability": 0.8955078125
                },
                {
                    "start": 568.03,
                    "end": 568.15,
                    "word": " it",
                    "probability": 0.89892578125
                },
                {
                    "start": 568.15,
                    "end": 568.29,
                    "word": " is",
                    "probability": 0.78515625
                },
                {
                    "start": 568.29,
                    "end": 568.57,
                    "word": " hard",
                    "probability": 0.798828125
                },
                {
                    "start": 568.57,
                    "end": 568.73,
                    "word": " to",
                    "probability": 0.904296875
                },
                {
                    "start": 568.73,
                    "end": 568.89,
                    "word": " get",
                    "probability": 0.89404296875
                },
                {
                    "start": 568.89,
                    "end": 569.01,
                    "word": " it",
                    "probability": 0.8935546875
                },
                {
                    "start": 569.01,
                    "end": 569.29,
                    "word": " right",
                    "probability": 0.88916015625
                },
                {
                    "start": 569.29,
                    "end": 569.89,
                    "word": ".",
                    "probability": 0.91943359375
                }
            ]
        },
        {
            "id": 71,
            "seek": 59785,
            "start": 569.89,
            "end": 576.15,
            "text": " As all of you in the crypto community probably know, people should enroll their own crypto.",
            "tokens": [
                50364,
                1018,
                439,
                295,
                291,
                294,
                264,
                17240,
                1768,
                1391,
                458,
                11,
                561,
                820,
                12266,
                641,
                1065,
                17240,
                13,
                50664
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18529211681174196,
            "compression_ratio": 1.632,
            "no_speech_prob": 0.01971435546875,
            "words": [
                {
                    "start": 569.89,
                    "end": 570.33,
                    "word": " As",
                    "probability": 0.78857421875
                },
                {
                    "start": 570.33,
                    "end": 570.77,
                    "word": " all",
                    "probability": 0.8798828125
                },
                {
                    "start": 570.77,
                    "end": 570.91,
                    "word": " of",
                    "probability": 0.87939453125
                },
                {
                    "start": 570.91,
                    "end": 571.07,
                    "word": " you",
                    "probability": 0.89208984375
                },
                {
                    "start": 571.07,
                    "end": 571.31,
                    "word": " in",
                    "probability": 0.89892578125
                },
                {
                    "start": 571.31,
                    "end": 571.47,
                    "word": " the",
                    "probability": 0.7978515625
                },
                {
                    "start": 571.47,
                    "end": 571.67,
                    "word": " crypto",
                    "probability": 0.7724609375
                },
                {
                    "start": 571.67,
                    "end": 572.13,
                    "word": " community",
                    "probability": 0.79638671875
                },
                {
                    "start": 572.13,
                    "end": 572.55,
                    "word": " probably",
                    "probability": 0.896484375
                },
                {
                    "start": 572.55,
                    "end": 573.01,
                    "word": " know",
                    "probability": 0.81689453125
                },
                {
                    "start": 573.01,
                    "end": 573.97,
                    "word": ",",
                    "probability": 0.76318359375
                },
                {
                    "start": 573.97,
                    "end": 574.49,
                    "word": " people",
                    "probability": 0.87060546875
                },
                {
                    "start": 574.49,
                    "end": 574.69,
                    "word": " should",
                    "probability": 0.9287109375
                },
                {
                    "start": 574.69,
                    "end": 575.01,
                    "word": " enroll",
                    "probability": 0.87109375
                },
                {
                    "start": 575.01,
                    "end": 575.23,
                    "word": " their",
                    "probability": 0.826171875
                },
                {
                    "start": 575.23,
                    "end": 575.41,
                    "word": " own",
                    "probability": 0.81689453125
                },
                {
                    "start": 575.41,
                    "end": 575.85,
                    "word": " crypto",
                    "probability": 0.7626953125
                },
                {
                    "start": 575.85,
                    "end": 576.15,
                    "word": ".",
                    "probability": 0.78466796875
                }
            ]
        },
        {
            "id": 72,
            "seek": 59785,
            "start": 576.57,
            "end": 582.23,
            "text": " And this is also a really good advice for differential privacy and their implementations.",
            "tokens": [
                50714,
                400,
                341,
                307,
                611,
                257,
                534,
                665,
                5192,
                337,
                15756,
                11427,
                293,
                641,
                4445,
                763,
                13,
                50964
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18529211681174196,
            "compression_ratio": 1.632,
            "no_speech_prob": 0.01971435546875,
            "words": [
                {
                    "start": 576.57,
                    "end": 576.73,
                    "word": " And",
                    "probability": 0.82373046875
                },
                {
                    "start": 576.73,
                    "end": 576.93,
                    "word": " this",
                    "probability": 0.828125
                },
                {
                    "start": 576.93,
                    "end": 577.07,
                    "word": " is",
                    "probability": 0.90869140625
                },
                {
                    "start": 577.07,
                    "end": 577.41,
                    "word": " also",
                    "probability": 0.806640625
                },
                {
                    "start": 577.71,
                    "end": 577.97,
                    "word": " a really",
                    "probability": 0.76904296875
                },
                {
                    "start": 577.97,
                    "end": 578.27,
                    "word": " good",
                    "probability": 0.88916015625
                },
                {
                    "start": 578.27,
                    "end": 578.79,
                    "word": " advice",
                    "probability": 0.78466796875
                },
                {
                    "start": 578.79,
                    "end": 579.05,
                    "word": " for",
                    "probability": 0.8916015625
                },
                {
                    "start": 579.05,
                    "end": 579.51,
                    "word": " differential",
                    "probability": 0.7314453125
                },
                {
                    "start": 579.51,
                    "end": 580.09,
                    "word": " privacy",
                    "probability": 0.64208984375
                },
                {
                    "start": 580.09,
                    "end": 580.59,
                    "word": " and",
                    "probability": 0.8564453125
                },
                {
                    "start": 580.59,
                    "end": 580.75,
                    "word": " their",
                    "probability": 0.8740234375
                },
                {
                    "start": 580.75,
                    "end": 581.63,
                    "word": " implementations",
                    "probability": 0.90185546875
                },
                {
                    "start": 581.63,
                    "end": 582.23,
                    "word": ".",
                    "probability": 0.9765625
                }
            ]
        },
        {
            "id": 73,
            "seek": 59785,
            "start": 582.43,
            "end": 586.51,
            "text": " Reusing a hardened implementation is generally a good idea.",
            "tokens": [
                51014,
                1300,
                7981,
                257,
                42605,
                11420,
                307,
                5101,
                257,
                665,
                1558,
                13,
                51164
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18529211681174196,
            "compression_ratio": 1.632,
            "no_speech_prob": 0.01971435546875,
            "words": [
                {
                    "start": 582.43,
                    "end": 582.83,
                    "word": " Reusing",
                    "probability": 0.71533203125
                },
                {
                    "start": 583.13,
                    "end": 583.39,
                    "word": " a hardened",
                    "probability": 0.544921875
                },
                {
                    "start": 583.39,
                    "end": 583.95,
                    "word": " implementation",
                    "probability": 0.73046875
                },
                {
                    "start": 583.95,
                    "end": 584.51,
                    "word": " is",
                    "probability": 0.904296875
                },
                {
                    "start": 584.51,
                    "end": 584.93,
                    "word": " generally",
                    "probability": 0.83544921875
                },
                {
                    "start": 585.31,
                    "end": 585.51,
                    "word": " a good",
                    "probability": 0.87841796875
                },
                {
                    "start": 585.51,
                    "end": 585.91,
                    "word": " idea",
                    "probability": 0.78955078125
                },
                {
                    "start": 585.91,
                    "end": 586.51,
                    "word": ".",
                    "probability": 0.9853515625
                }
            ]
        },
        {
            "id": 74,
            "seek": 59785,
            "start": 586.89,
            "end": 594.15,
            "text": " We open source the technology, and you can read more of the articles that were written back then.",
            "tokens": [
                51214,
                492,
                1269,
                4009,
                264,
                2899,
                11,
                293,
                291,
                393,
                1401,
                544,
                295,
                264,
                11290,
                300,
                645,
                3720,
                646,
                550,
                13,
                51564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18529211681174196,
            "compression_ratio": 1.632,
            "no_speech_prob": 0.01971435546875,
            "words": [
                {
                    "start": 586.89,
                    "end": 587.11,
                    "word": " We",
                    "probability": 0.8916015625
                },
                {
                    "start": 587.11,
                    "end": 587.41,
                    "word": " open",
                    "probability": 0.7607421875
                },
                {
                    "start": 587.41,
                    "end": 587.77,
                    "word": " source",
                    "probability": 0.469970703125
                },
                {
                    "start": 587.77,
                    "end": 588.97,
                    "word": " the",
                    "probability": 0.7490234375
                },
                {
                    "start": 588.97,
                    "end": 589.43,
                    "word": " technology",
                    "probability": 0.9033203125
                },
                {
                    "start": 589.43,
                    "end": 589.55,
                    "word": ",",
                    "probability": 0.222900390625
                },
                {
                    "start": 589.55,
                    "end": 589.63,
                    "word": " and",
                    "probability": 0.88818359375
                },
                {
                    "start": 589.63,
                    "end": 589.77,
                    "word": " you",
                    "probability": 0.89892578125
                },
                {
                    "start": 589.77,
                    "end": 589.91,
                    "word": " can",
                    "probability": 0.90380859375
                },
                {
                    "start": 589.91,
                    "end": 590.09,
                    "word": " read",
                    "probability": 0.84033203125
                },
                {
                    "start": 590.09,
                    "end": 590.43,
                    "word": " more",
                    "probability": 0.8662109375
                },
                {
                    "start": 590.43,
                    "end": 590.61,
                    "word": " of",
                    "probability": 0.87548828125
                },
                {
                    "start": 590.61,
                    "end": 590.75,
                    "word": " the",
                    "probability": 0.8154296875
                },
                {
                    "start": 590.75,
                    "end": 591.19,
                    "word": " articles",
                    "probability": 0.84033203125
                },
                {
                    "start": 591.19,
                    "end": 592.49,
                    "word": " that",
                    "probability": 0.8466796875
                },
                {
                    "start": 592.49,
                    "end": 592.75,
                    "word": " were",
                    "probability": 0.86376953125
                },
                {
                    "start": 592.75,
                    "end": 593.23,
                    "word": " written",
                    "probability": 0.85107421875
                },
                {
                    "start": 593.23,
                    "end": 593.55,
                    "word": " back",
                    "probability": 0.82568359375
                },
                {
                    "start": 593.55,
                    "end": 593.85,
                    "word": " then",
                    "probability": 0.72412109375
                },
                {
                    "start": 593.85,
                    "end": 594.15,
                    "word": ".",
                    "probability": 0.98583984375
                }
            ]
        },
        {
            "id": 75,
            "seek": 59785,
            "start": 595.21,
            "end": 599.51,
            "text": " I'll now give you some insights as to how we've used the technology.",
            "tokens": [
                51614,
                286,
                603,
                586,
                976,
                291,
                512,
                14310,
                382,
                281,
                577,
                321,
                600,
                1143,
                264,
                2899,
                13,
                51814
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18529211681174196,
            "compression_ratio": 1.632,
            "no_speech_prob": 0.01971435546875,
            "words": [
                {
                    "start": 595.21,
                    "end": 595.41,
                    "word": " I'll",
                    "probability": 0.600341796875
                },
                {
                    "start": 595.41,
                    "end": 595.55,
                    "word": " now",
                    "probability": 0.79833984375
                },
                {
                    "start": 595.55,
                    "end": 595.75,
                    "word": " give",
                    "probability": 0.83935546875
                },
                {
                    "start": 595.75,
                    "end": 595.95,
                    "word": " you",
                    "probability": 0.89111328125
                },
                {
                    "start": 595.95,
                    "end": 596.41,
                    "word": " some",
                    "probability": 0.81591796875
                },
                {
                    "start": 596.41,
                    "end": 596.95,
                    "word": " insights",
                    "probability": 0.81689453125
                },
                {
                    "start": 596.95,
                    "end": 597.27,
                    "word": " as",
                    "probability": 0.8896484375
                },
                {
                    "start": 597.27,
                    "end": 597.39,
                    "word": " to",
                    "probability": 0.9033203125
                },
                {
                    "start": 597.39,
                    "end": 597.63,
                    "word": " how",
                    "probability": 0.892578125
                },
                {
                    "start": 597.63,
                    "end": 597.91,
                    "word": " we've",
                    "probability": 0.837158203125
                },
                {
                    "start": 597.91,
                    "end": 598.11,
                    "word": " used",
                    "probability": 0.794921875
                },
                {
                    "start": 598.11,
                    "end": 598.33,
                    "word": " the",
                    "probability": 0.67236328125
                },
                {
                    "start": 598.33,
                    "end": 599.27,
                    "word": " technology",
                    "probability": 0.9033203125
                },
                {
                    "start": 599.27,
                    "end": 599.51,
                    "word": ".",
                    "probability": 0.9775390625
                }
            ]
        },
        {
            "id": 76,
            "seek": 62557,
            "start": 601.35,
            "end": 605.13,
            "text": " You're probably aware of the Google Mobility Reports.",
            "tokens": [
                50414,
                509,
                434,
                1391,
                3650,
                295,
                264,
                3329,
                37920,
                1140,
                45910,
                13,
                50564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20331325049859933,
            "compression_ratio": 1.5481171548117154,
            "no_speech_prob": 0.0034275054931640625,
            "words": [
                {
                    "start": 601.35,
                    "end": 602.07,
                    "word": " You're",
                    "probability": 0.83154296875
                },
                {
                    "start": 602.07,
                    "end": 602.29,
                    "word": " probably",
                    "probability": 0.92333984375
                },
                {
                    "start": 602.29,
                    "end": 602.69,
                    "word": " aware",
                    "probability": 0.7998046875
                },
                {
                    "start": 602.69,
                    "end": 603.25,
                    "word": " of",
                    "probability": 0.90234375
                },
                {
                    "start": 603.25,
                    "end": 603.45,
                    "word": " the",
                    "probability": 0.80859375
                },
                {
                    "start": 603.45,
                    "end": 603.73,
                    "word": " Google",
                    "probability": 0.84375
                },
                {
                    "start": 603.73,
                    "end": 604.27,
                    "word": " Mobility",
                    "probability": 0.7352294921875
                },
                {
                    "start": 604.27,
                    "end": 604.77,
                    "word": " Reports",
                    "probability": 0.6005859375
                },
                {
                    "start": 604.77,
                    "end": 605.13,
                    "word": ".",
                    "probability": 0.96240234375
                }
            ]
        },
        {
            "id": 77,
            "seek": 62557,
            "start": 605.13,
            "end": 612.43,
            "text": " These Mobility Reports were a response to the emerging crisis back in March around COVID-19.",
            "tokens": [
                50614,
                1981,
                37920,
                1140,
                45910,
                645,
                257,
                4134,
                281,
                264,
                14989,
                5869,
                646,
                294,
                6129,
                926,
                4566,
                12,
                3405,
                13,
                50964
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20331325049859933,
            "compression_ratio": 1.5481171548117154,
            "no_speech_prob": 0.0034275054931640625,
            "words": [
                {
                    "start": 605.13,
                    "end": 605.35,
                    "word": " These",
                    "probability": 0.8642578125
                },
                {
                    "start": 605.35,
                    "end": 605.85,
                    "word": " Mobility",
                    "probability": 0.7335205078125
                },
                {
                    "start": 605.85,
                    "end": 606.25,
                    "word": " Reports",
                    "probability": 0.7978515625
                },
                {
                    "start": 606.25,
                    "end": 606.67,
                    "word": " were",
                    "probability": 0.873046875
                },
                {
                    "start": 606.97,
                    "end": 607.51,
                    "word": " a response",
                    "probability": 0.89208984375
                },
                {
                    "start": 607.51,
                    "end": 608.05,
                    "word": " to",
                    "probability": 0.89892578125
                },
                {
                    "start": 608.05,
                    "end": 608.37,
                    "word": " the",
                    "probability": 0.80859375
                },
                {
                    "start": 608.37,
                    "end": 608.93,
                    "word": " emerging",
                    "probability": 0.7138671875
                },
                {
                    "start": 608.93,
                    "end": 609.41,
                    "word": " crisis",
                    "probability": 0.60693359375
                },
                {
                    "start": 609.41,
                    "end": 609.81,
                    "word": " back",
                    "probability": 0.82763671875
                },
                {
                    "start": 609.81,
                    "end": 610.09,
                    "word": " in",
                    "probability": 0.91357421875
                },
                {
                    "start": 610.09,
                    "end": 610.45,
                    "word": " March",
                    "probability": 0.8798828125
                },
                {
                    "start": 610.45,
                    "end": 611.15,
                    "word": " around",
                    "probability": 0.74365234375
                },
                {
                    "start": 611.15,
                    "end": 611.51,
                    "word": " COVID",
                    "probability": 0.88818359375
                },
                {
                    "start": 611.51,
                    "end": 611.93,
                    "word": "-19",
                    "probability": 0.985107421875
                },
                {
                    "start": 611.93,
                    "end": 612.43,
                    "word": ".",
                    "probability": 0.98095703125
                }
            ]
        },
        {
            "id": 78,
            "seek": 62557,
            "start": 612.43,
            "end": 620.89,
            "text": " They provide policymakers with an idea of mobility trends at a city and state level for different categories.",
            "tokens": [
                51014,
                814,
                2893,
                47325,
                365,
                364,
                1558,
                295,
                16199,
                13892,
                412,
                257,
                2307,
                293,
                1785,
                1496,
                337,
                819,
                10479,
                13,
                51364
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20331325049859933,
            "compression_ratio": 1.5481171548117154,
            "no_speech_prob": 0.0034275054931640625,
            "words": [
                {
                    "start": 612.43,
                    "end": 612.65,
                    "word": " They",
                    "probability": 0.72509765625
                },
                {
                    "start": 612.65,
                    "end": 613.03,
                    "word": " provide",
                    "probability": 0.88623046875
                },
                {
                    "start": 613.03,
                    "end": 613.67,
                    "word": " policymakers",
                    "probability": 0.45654296875
                },
                {
                    "start": 613.67,
                    "end": 614.17,
                    "word": " with",
                    "probability": 0.81396484375
                },
                {
                    "start": 614.17,
                    "end": 614.37,
                    "word": " an",
                    "probability": 0.90869140625
                },
                {
                    "start": 614.37,
                    "end": 614.77,
                    "word": " idea",
                    "probability": 0.77685546875
                },
                {
                    "start": 614.77,
                    "end": 615.07,
                    "word": " of",
                    "probability": 0.90283203125
                },
                {
                    "start": 615.07,
                    "end": 615.45,
                    "word": " mobility",
                    "probability": 0.828125
                },
                {
                    "start": 615.45,
                    "end": 615.99,
                    "word": " trends",
                    "probability": 0.70849609375
                },
                {
                    "start": 615.99,
                    "end": 616.51,
                    "word": " at",
                    "probability": 0.8876953125
                },
                {
                    "start": 616.87,
                    "end": 617.11,
                    "word": " a city",
                    "probability": 0.81103515625
                },
                {
                    "start": 617.11,
                    "end": 617.57,
                    "word": " and",
                    "probability": 0.88232421875
                },
                {
                    "start": 617.57,
                    "end": 617.91,
                    "word": " state",
                    "probability": 0.86181640625
                },
                {
                    "start": 617.91,
                    "end": 618.31,
                    "word": " level",
                    "probability": 0.798828125
                },
                {
                    "start": 618.31,
                    "end": 619.27,
                    "word": " for",
                    "probability": 0.85009765625
                },
                {
                    "start": 619.27,
                    "end": 619.59,
                    "word": " different",
                    "probability": 0.68603515625
                },
                {
                    "start": 619.59,
                    "end": 620.17,
                    "word": " categories",
                    "probability": 0.77685546875
                },
                {
                    "start": 620.17,
                    "end": 620.89,
                    "word": ".",
                    "probability": 0.9970703125
                }
            ]
        },
        {
            "id": 79,
            "seek": 62557,
            "start": 621.03,
            "end": 627.61,
            "text": " Before I continue going into the Mobility Reports, let me talk to you a little bit about location data at Google.",
            "tokens": [
                51414,
                4546,
                286,
                2354,
                516,
                666,
                264,
                37920,
                1140,
                45910,
                11,
                718,
                385,
                751,
                281,
                291,
                257,
                707,
                857,
                466,
                4914,
                1412,
                412,
                3329,
                13,
                51714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20331325049859933,
            "compression_ratio": 1.5481171548117154,
            "no_speech_prob": 0.0034275054931640625,
            "words": [
                {
                    "start": 621.03,
                    "end": 621.25,
                    "word": " Before",
                    "probability": 0.83935546875
                },
                {
                    "start": 621.25,
                    "end": 621.43,
                    "word": " I",
                    "probability": 0.9892578125
                },
                {
                    "start": 621.43,
                    "end": 621.69,
                    "word": " continue",
                    "probability": 0.802734375
                },
                {
                    "start": 621.69,
                    "end": 622.03,
                    "word": " going",
                    "probability": 0.81298828125
                },
                {
                    "start": 622.03,
                    "end": 622.25,
                    "word": " into",
                    "probability": 0.728515625
                },
                {
                    "start": 622.25,
                    "end": 622.45,
                    "word": " the",
                    "probability": 0.80859375
                },
                {
                    "start": 622.45,
                    "end": 622.87,
                    "word": " Mobility",
                    "probability": 0.829345703125
                },
                {
                    "start": 622.87,
                    "end": 623.33,
                    "word": " Reports",
                    "probability": 0.78662109375
                },
                {
                    "start": 623.33,
                    "end": 623.63,
                    "word": ",",
                    "probability": 0.9375
                },
                {
                    "start": 623.63,
                    "end": 623.77,
                    "word": " let",
                    "probability": 0.89453125
                },
                {
                    "start": 623.77,
                    "end": 623.87,
                    "word": " me",
                    "probability": 0.8984375
                },
                {
                    "start": 623.87,
                    "end": 624.27,
                    "word": " talk",
                    "probability": 0.8046875
                },
                {
                    "start": 624.27,
                    "end": 624.41,
                    "word": " to",
                    "probability": 0.90673828125
                },
                {
                    "start": 624.41,
                    "end": 624.59,
                    "word": " you",
                    "probability": 0.8955078125
                },
                {
                    "start": 624.79,
                    "end": 624.95,
                    "word": " a little",
                    "probability": 0.8017578125
                },
                {
                    "start": 624.95,
                    "end": 625.19,
                    "word": " bit",
                    "probability": 0.8974609375
                },
                {
                    "start": 625.19,
                    "end": 625.45,
                    "word": " about",
                    "probability": 0.73779296875
                },
                {
                    "start": 625.45,
                    "end": 625.93,
                    "word": " location",
                    "probability": 0.74462890625
                },
                {
                    "start": 625.93,
                    "end": 626.27,
                    "word": " data",
                    "probability": 0.80126953125
                },
                {
                    "start": 626.27,
                    "end": 626.55,
                    "word": " at",
                    "probability": 0.90185546875
                },
                {
                    "start": 626.55,
                    "end": 626.89,
                    "word": " Google",
                    "probability": 0.86181640625
                },
                {
                    "start": 626.89,
                    "end": 627.61,
                    "word": ".",
                    "probability": 0.9970703125
                }
            ]
        },
        {
            "id": 80,
            "seek": 65465,
            "start": 629.43,
            "end": 635.27,
            "text": " Location data is an optional account feature that says where you go in your private timeline.",
            "tokens": [
                50414,
                12859,
                399,
                1412,
                307,
                364,
                17312,
                2696,
                4111,
                300,
                1619,
                689,
                291,
                352,
                294,
                428,
                4551,
                12933,
                13,
                50714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19615964286298637,
            "compression_ratio": 1.563265306122449,
            "no_speech_prob": 0.00490570068359375,
            "words": [
                {
                    "start": 629.43,
                    "end": 630.11,
                    "word": " Location",
                    "probability": 0.689208984375
                },
                {
                    "start": 630.11,
                    "end": 630.41,
                    "word": " data",
                    "probability": 0.69189453125
                },
                {
                    "start": 630.41,
                    "end": 630.81,
                    "word": " is",
                    "probability": 0.89697265625
                },
                {
                    "start": 630.81,
                    "end": 630.99,
                    "word": " an",
                    "probability": 0.9052734375
                },
                {
                    "start": 630.99,
                    "end": 631.33,
                    "word": " optional",
                    "probability": 0.79638671875
                },
                {
                    "start": 631.33,
                    "end": 631.67,
                    "word": " account",
                    "probability": 0.75
                },
                {
                    "start": 631.67,
                    "end": 632.11,
                    "word": " feature",
                    "probability": 0.88623046875
                },
                {
                    "start": 632.11,
                    "end": 632.45,
                    "word": " that",
                    "probability": 0.84619140625
                },
                {
                    "start": 632.45,
                    "end": 632.67,
                    "word": " says",
                    "probability": 0.4560546875
                },
                {
                    "start": 632.67,
                    "end": 632.93,
                    "word": " where",
                    "probability": 0.9111328125
                },
                {
                    "start": 632.93,
                    "end": 633.13,
                    "word": " you",
                    "probability": 0.892578125
                },
                {
                    "start": 633.13,
                    "end": 633.39,
                    "word": " go",
                    "probability": 0.8955078125
                },
                {
                    "start": 633.39,
                    "end": 633.67,
                    "word": " in",
                    "probability": 0.8974609375
                },
                {
                    "start": 633.67,
                    "end": 633.85,
                    "word": " your",
                    "probability": 0.8037109375
                },
                {
                    "start": 633.85,
                    "end": 634.29,
                    "word": " private",
                    "probability": 0.71533203125
                },
                {
                    "start": 634.29,
                    "end": 634.87,
                    "word": " timeline",
                    "probability": 0.8759765625
                },
                {
                    "start": 634.87,
                    "end": 635.27,
                    "word": ".",
                    "probability": 0.94970703125
                }
            ]
        },
        {
            "id": 81,
            "seek": 65465,
            "start": 635.27,
            "end": 639.39,
            "text": " It's off by default and can be paused or unpaused anytime.",
            "tokens": [
                50764,
                467,
                311,
                766,
                538,
                7576,
                293,
                393,
                312,
                46860,
                420,
                517,
                4306,
                4717,
                13038,
                13,
                50914
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19615964286298637,
            "compression_ratio": 1.563265306122449,
            "no_speech_prob": 0.00490570068359375,
            "words": [
                {
                    "start": 635.27,
                    "end": 635.67,
                    "word": " It's",
                    "probability": 0.93994140625
                },
                {
                    "start": 635.67,
                    "end": 635.89,
                    "word": " off",
                    "probability": 0.79296875
                },
                {
                    "start": 635.89,
                    "end": 636.09,
                    "word": " by",
                    "probability": 0.90673828125
                },
                {
                    "start": 636.49,
                    "end": 636.75,
                    "word": " default and",
                    "probability": 0.79638671875
                },
                {
                    "start": 636.75,
                    "end": 636.97,
                    "word": " can",
                    "probability": 0.896484375
                },
                {
                    "start": 636.97,
                    "end": 637.13,
                    "word": " be",
                    "probability": 0.90869140625
                },
                {
                    "start": 637.13,
                    "end": 637.49,
                    "word": " paused",
                    "probability": 0.84228515625
                },
                {
                    "start": 637.49,
                    "end": 637.75,
                    "word": " or",
                    "probability": 0.89892578125
                },
                {
                    "start": 637.75,
                    "end": 638.15,
                    "word": " unpaused",
                    "probability": 0.7511393229166666
                },
                {
                    "start": 638.15,
                    "end": 638.71,
                    "word": " anytime",
                    "probability": 0.40283203125
                },
                {
                    "start": 638.71,
                    "end": 639.39,
                    "word": ".",
                    "probability": 0.98681640625
                }
            ]
        },
        {
            "id": 82,
            "seek": 65465,
            "start": 639.99,
            "end": 644.39,
            "text": " From a product perspective, it's available on Android and iOS.",
            "tokens": [
                50964,
                3358,
                257,
                1674,
                4585,
                11,
                309,
                311,
                2435,
                322,
                8853,
                293,
                17430,
                13,
                51164
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19615964286298637,
            "compression_ratio": 1.563265306122449,
            "no_speech_prob": 0.00490570068359375,
            "words": [
                {
                    "start": 639.99,
                    "end": 640.23,
                    "word": " From",
                    "probability": 0.7021484375
                },
                {
                    "start": 641.09,
                    "end": 641.27,
                    "word": " a product",
                    "probability": 0.6142578125
                },
                {
                    "start": 641.27,
                    "end": 641.77,
                    "word": " perspective",
                    "probability": 0.787109375
                },
                {
                    "start": 641.77,
                    "end": 641.91,
                    "word": ",",
                    "probability": 0.90771484375
                },
                {
                    "start": 641.91,
                    "end": 642.13,
                    "word": " it's",
                    "probability": 0.94677734375
                },
                {
                    "start": 642.13,
                    "end": 642.47,
                    "word": " available",
                    "probability": 0.73779296875
                },
                {
                    "start": 642.47,
                    "end": 642.75,
                    "word": " on",
                    "probability": 0.80078125
                },
                {
                    "start": 642.75,
                    "end": 643.09,
                    "word": " Android",
                    "probability": 0.9287109375
                },
                {
                    "start": 643.09,
                    "end": 643.37,
                    "word": " and",
                    "probability": 0.90283203125
                },
                {
                    "start": 643.37,
                    "end": 643.71,
                    "word": " iOS",
                    "probability": 0.9248046875
                },
                {
                    "start": 643.71,
                    "end": 644.39,
                    "word": ".",
                    "probability": 0.9873046875
                }
            ]
        },
        {
            "id": 83,
            "seek": 65465,
            "start": 644.99,
            "end": 656.69,
            "text": " It helps users remember where they have been and gives users personalized information such as better restaurant recommendations, suggestions for a faster commute, etc.",
            "tokens": [
                51214,
                467,
                3665,
                5022,
                1604,
                689,
                436,
                362,
                668,
                293,
                2709,
                5022,
                28415,
                1589,
                1270,
                382,
                1101,
                6383,
                10434,
                11,
                13396,
                337,
                257,
                4663,
                36750,
                11,
                5183,
                13,
                51764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19615964286298637,
            "compression_ratio": 1.563265306122449,
            "no_speech_prob": 0.00490570068359375,
            "words": [
                {
                    "start": 644.99,
                    "end": 645.23,
                    "word": " It",
                    "probability": 0.90087890625
                },
                {
                    "start": 645.23,
                    "end": 645.57,
                    "word": " helps",
                    "probability": 0.70654296875
                },
                {
                    "start": 645.57,
                    "end": 646.17,
                    "word": " users",
                    "probability": 0.82568359375
                },
                {
                    "start": 646.17,
                    "end": 646.91,
                    "word": " remember",
                    "probability": 0.76123046875
                },
                {
                    "start": 646.91,
                    "end": 647.19,
                    "word": " where",
                    "probability": 0.908203125
                },
                {
                    "start": 647.19,
                    "end": 647.35,
                    "word": " they",
                    "probability": 0.73876953125
                },
                {
                    "start": 647.35,
                    "end": 647.57,
                    "word": " have",
                    "probability": 0.86328125
                },
                {
                    "start": 647.57,
                    "end": 647.85,
                    "word": " been",
                    "probability": 0.8251953125
                },
                {
                    "start": 647.85,
                    "end": 648.13,
                    "word": " and",
                    "probability": 0.828125
                },
                {
                    "start": 648.13,
                    "end": 648.33,
                    "word": " gives",
                    "probability": 0.74560546875
                },
                {
                    "start": 648.33,
                    "end": 649.71,
                    "word": " users",
                    "probability": 0.8076171875
                },
                {
                    "start": 649.71,
                    "end": 650.19,
                    "word": " personalized",
                    "probability": 0.8994140625
                },
                {
                    "start": 650.19,
                    "end": 650.95,
                    "word": " information",
                    "probability": 0.73779296875
                },
                {
                    "start": 650.95,
                    "end": 651.29,
                    "word": " such",
                    "probability": 0.69287109375
                },
                {
                    "start": 651.29,
                    "end": 651.63,
                    "word": " as",
                    "probability": 0.90478515625
                },
                {
                    "start": 651.63,
                    "end": 652.53,
                    "word": " better",
                    "probability": 0.798828125
                },
                {
                    "start": 652.53,
                    "end": 652.95,
                    "word": " restaurant",
                    "probability": 0.80126953125
                },
                {
                    "start": 652.95,
                    "end": 653.63,
                    "word": " recommendations",
                    "probability": 0.64892578125
                },
                {
                    "start": 653.63,
                    "end": 654.31,
                    "word": ",",
                    "probability": 0.98681640625
                },
                {
                    "start": 654.31,
                    "end": 654.69,
                    "word": " suggestions",
                    "probability": 0.75927734375
                },
                {
                    "start": 654.69,
                    "end": 654.95,
                    "word": " for",
                    "probability": 0.88916015625
                },
                {
                    "start": 655.07,
                    "end": 655.31,
                    "word": " a faster",
                    "probability": 0.69482421875
                },
                {
                    "start": 655.31,
                    "end": 655.79,
                    "word": " commute",
                    "probability": 0.92333984375
                },
                {
                    "start": 655.79,
                    "end": 655.91,
                    "word": ",",
                    "probability": 0.96435546875
                },
                {
                    "start": 655.91,
                    "end": 656.17,
                    "word": " etc",
                    "probability": 0.426025390625
                },
                {
                    "start": 656.17,
                    "end": 656.69,
                    "word": ".",
                    "probability": 0.98486328125
                }
            ]
        },
        {
            "id": 84,
            "seek": 68319,
            "start": 656.69,
            "end": 667.09,
            "text": " Users can view, edit, and delete some or all of their timeline at any time, as well as set up auto deletion after three or 18 months.",
            "tokens": [
                50364,
                47092,
                393,
                1910,
                11,
                8129,
                11,
                293,
                12097,
                512,
                420,
                439,
                295,
                641,
                12933,
                412,
                604,
                565,
                11,
                382,
                731,
                382,
                992,
                493,
                8399,
                1103,
                302,
                313,
                934,
                1045,
                420,
                2443,
                2493,
                13,
                50864
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20135308786765815,
            "compression_ratio": 1.5367965367965368,
            "no_speech_prob": 0.00225067138671875,
            "words": [
                {
                    "start": 656.69,
                    "end": 657.27,
                    "word": " Users",
                    "probability": 0.5546875
                },
                {
                    "start": 657.27,
                    "end": 657.61,
                    "word": " can",
                    "probability": 0.89501953125
                },
                {
                    "start": 657.61,
                    "end": 657.89,
                    "word": " view",
                    "probability": 0.84716796875
                },
                {
                    "start": 657.89,
                    "end": 658.19,
                    "word": ",",
                    "probability": 0.94140625
                },
                {
                    "start": 658.19,
                    "end": 658.41,
                    "word": " edit",
                    "probability": 0.84033203125
                },
                {
                    "start": 658.41,
                    "end": 658.57,
                    "word": ",",
                    "probability": 0.54931640625
                },
                {
                    "start": 658.57,
                    "end": 658.67,
                    "word": " and",
                    "probability": 0.8876953125
                },
                {
                    "start": 658.67,
                    "end": 658.93,
                    "word": " delete",
                    "probability": 0.7607421875
                },
                {
                    "start": 658.93,
                    "end": 659.31,
                    "word": " some",
                    "probability": 0.73583984375
                },
                {
                    "start": 659.31,
                    "end": 659.59,
                    "word": " or",
                    "probability": 0.85888671875
                },
                {
                    "start": 659.59,
                    "end": 661.03,
                    "word": " all",
                    "probability": 0.90380859375
                },
                {
                    "start": 661.03,
                    "end": 661.19,
                    "word": " of",
                    "probability": 0.85986328125
                },
                {
                    "start": 661.19,
                    "end": 661.29,
                    "word": " their",
                    "probability": 0.84912109375
                },
                {
                    "start": 661.29,
                    "end": 661.87,
                    "word": " timeline",
                    "probability": 0.7919921875
                },
                {
                    "start": 661.87,
                    "end": 662.95,
                    "word": " at",
                    "probability": 0.888671875
                },
                {
                    "start": 662.95,
                    "end": 663.17,
                    "word": " any",
                    "probability": 0.765625
                },
                {
                    "start": 663.17,
                    "end": 663.45,
                    "word": " time",
                    "probability": 0.80224609375
                },
                {
                    "start": 663.45,
                    "end": 663.61,
                    "word": ",",
                    "probability": 0.7880859375
                },
                {
                    "start": 663.61,
                    "end": 663.83,
                    "word": " as",
                    "probability": 0.9033203125
                },
                {
                    "start": 663.83,
                    "end": 663.99,
                    "word": " well",
                    "probability": 0.8740234375
                },
                {
                    "start": 663.99,
                    "end": 664.17,
                    "word": " as",
                    "probability": 0.9013671875
                },
                {
                    "start": 664.17,
                    "end": 664.31,
                    "word": " set",
                    "probability": 0.76953125
                },
                {
                    "start": 664.31,
                    "end": 664.47,
                    "word": " up",
                    "probability": 0.908203125
                },
                {
                    "start": 664.47,
                    "end": 664.69,
                    "word": " auto",
                    "probability": 0.5205078125
                },
                {
                    "start": 664.69,
                    "end": 665.09,
                    "word": " deletion",
                    "probability": 0.79833984375
                },
                {
                    "start": 665.09,
                    "end": 665.35,
                    "word": " after",
                    "probability": 0.74267578125
                },
                {
                    "start": 665.35,
                    "end": 665.77,
                    "word": " three",
                    "probability": 0.5625
                },
                {
                    "start": 665.77,
                    "end": 665.99,
                    "word": " or",
                    "probability": 0.90966796875
                },
                {
                    "start": 665.99,
                    "end": 666.35,
                    "word": " 18",
                    "probability": 0.88134765625
                },
                {
                    "start": 666.35,
                    "end": 666.79,
                    "word": " months",
                    "probability": 0.73828125
                },
                {
                    "start": 666.79,
                    "end": 667.09,
                    "word": ".",
                    "probability": 0.99560546875
                }
            ]
        },
        {
            "id": 85,
            "seek": 68319,
            "start": 667.09,
            "end": 672.27,
            "text": " More recently, we announced that auto deletion is now automatic for new accounts.",
            "tokens": [
                50914,
                5048,
                3938,
                11,
                321,
                7548,
                300,
                8399,
                1103,
                302,
                313,
                307,
                586,
                12509,
                337,
                777,
                9402,
                13,
                51114
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20135308786765815,
            "compression_ratio": 1.5367965367965368,
            "no_speech_prob": 0.00225067138671875,
            "words": [
                {
                    "start": 667.09,
                    "end": 667.37,
                    "word": " More",
                    "probability": 0.80078125
                },
                {
                    "start": 667.37,
                    "end": 667.79,
                    "word": " recently",
                    "probability": 0.8056640625
                },
                {
                    "start": 667.79,
                    "end": 668.09,
                    "word": ",",
                    "probability": 0.8779296875
                },
                {
                    "start": 668.09,
                    "end": 668.19,
                    "word": " we",
                    "probability": 0.83447265625
                },
                {
                    "start": 668.19,
                    "end": 668.59,
                    "word": " announced",
                    "probability": 0.791015625
                },
                {
                    "start": 668.59,
                    "end": 668.79,
                    "word": " that",
                    "probability": 0.84716796875
                },
                {
                    "start": 668.79,
                    "end": 668.99,
                    "word": " auto",
                    "probability": 0.771484375
                },
                {
                    "start": 668.99,
                    "end": 669.43,
                    "word": " deletion",
                    "probability": 0.9275716145833334
                },
                {
                    "start": 669.43,
                    "end": 669.79,
                    "word": " is",
                    "probability": 0.90625
                },
                {
                    "start": 669.79,
                    "end": 670.17,
                    "word": " now",
                    "probability": 0.8896484375
                },
                {
                    "start": 670.17,
                    "end": 670.71,
                    "word": " automatic",
                    "probability": 0.76904296875
                },
                {
                    "start": 670.71,
                    "end": 671.05,
                    "word": " for",
                    "probability": 0.8984375
                },
                {
                    "start": 671.05,
                    "end": 671.27,
                    "word": " new",
                    "probability": 0.79345703125
                },
                {
                    "start": 671.27,
                    "end": 671.75,
                    "word": " accounts",
                    "probability": 0.7470703125
                },
                {
                    "start": 671.75,
                    "end": 672.27,
                    "word": ".",
                    "probability": 0.85009765625
                }
            ]
        },
        {
            "id": 86,
            "seek": 68319,
            "start": 672.59,
            "end": 681.11,
            "text": " And if you're curious as to how it looks like, you can access it as it's been shown on this GIF.",
            "tokens": [
                51164,
                400,
                498,
                291,
                434,
                6369,
                382,
                281,
                577,
                309,
                1542,
                411,
                11,
                291,
                393,
                2105,
                309,
                382,
                309,
                311,
                668,
                4898,
                322,
                341,
                460,
                12775,
                13,
                51564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20135308786765815,
            "compression_ratio": 1.5367965367965368,
            "no_speech_prob": 0.00225067138671875,
            "words": [
                {
                    "start": 672.59,
                    "end": 672.71,
                    "word": " And",
                    "probability": 0.826171875
                },
                {
                    "start": 672.71,
                    "end": 672.83,
                    "word": " if",
                    "probability": 0.89990234375
                },
                {
                    "start": 672.83,
                    "end": 673.07,
                    "word": " you're",
                    "probability": 0.907958984375
                },
                {
                    "start": 673.07,
                    "end": 673.45,
                    "word": " curious",
                    "probability": 0.78662109375
                },
                {
                    "start": 673.45,
                    "end": 673.71,
                    "word": " as",
                    "probability": 0.8828125
                },
                {
                    "start": 673.71,
                    "end": 673.81,
                    "word": " to",
                    "probability": 0.90380859375
                },
                {
                    "start": 673.81,
                    "end": 674.01,
                    "word": " how",
                    "probability": 0.90576171875
                },
                {
                    "start": 674.01,
                    "end": 674.19,
                    "word": " it",
                    "probability": 0.90625
                },
                {
                    "start": 674.19,
                    "end": 674.41,
                    "word": " looks",
                    "probability": 0.78369140625
                },
                {
                    "start": 674.41,
                    "end": 674.65,
                    "word": " like",
                    "probability": 0.8759765625
                },
                {
                    "start": 674.65,
                    "end": 674.69,
                    "word": ",",
                    "probability": 0.80322265625
                },
                {
                    "start": 674.69,
                    "end": 674.79,
                    "word": " you",
                    "probability": 0.89208984375
                },
                {
                    "start": 674.79,
                    "end": 674.93,
                    "word": " can",
                    "probability": 0.892578125
                },
                {
                    "start": 674.93,
                    "end": 675.29,
                    "word": " access",
                    "probability": 0.8046875
                },
                {
                    "start": 675.29,
                    "end": 675.71,
                    "word": " it",
                    "probability": 0.8173828125
                },
                {
                    "start": 675.71,
                    "end": 676.47,
                    "word": " as",
                    "probability": 0.6748046875
                },
                {
                    "start": 676.47,
                    "end": 676.91,
                    "word": " it's",
                    "probability": 0.94775390625
                },
                {
                    "start": 676.91,
                    "end": 677.07,
                    "word": " been",
                    "probability": 0.68896484375
                },
                {
                    "start": 677.07,
                    "end": 677.37,
                    "word": " shown",
                    "probability": 0.8740234375
                },
                {
                    "start": 677.37,
                    "end": 679.15,
                    "word": " on",
                    "probability": 0.8994140625
                },
                {
                    "start": 679.15,
                    "end": 679.51,
                    "word": " this",
                    "probability": 0.87353515625
                },
                {
                    "start": 679.51,
                    "end": 680.59,
                    "word": " GIF",
                    "probability": 0.7548828125
                },
                {
                    "start": 680.59,
                    "end": 681.11,
                    "word": ".",
                    "probability": 0.99609375
                }
            ]
        },
        {
            "id": 87,
            "seek": 68319,
            "start": 682.69,
            "end": 685.23,
            "text": " Now, let me talk to you about the reports.",
            "tokens": [
                51664,
                823,
                11,
                718,
                385,
                751,
                281,
                291,
                466,
                264,
                7122,
                13,
                51764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20135308786765815,
            "compression_ratio": 1.5367965367965368,
            "no_speech_prob": 0.00225067138671875,
            "words": [
                {
                    "start": 682.69,
                    "end": 683.21,
                    "word": " Now",
                    "probability": 0.88671875
                },
                {
                    "start": 683.21,
                    "end": 683.39,
                    "word": ",",
                    "probability": 0.802734375
                },
                {
                    "start": 683.39,
                    "end": 683.61,
                    "word": " let",
                    "probability": 0.90478515625
                },
                {
                    "start": 683.61,
                    "end": 683.69,
                    "word": " me",
                    "probability": 0.904296875
                },
                {
                    "start": 683.69,
                    "end": 683.85,
                    "word": " talk",
                    "probability": 0.8154296875
                },
                {
                    "start": 683.85,
                    "end": 683.97,
                    "word": " to",
                    "probability": 0.90576171875
                },
                {
                    "start": 683.97,
                    "end": 684.13,
                    "word": " you",
                    "probability": 0.89599609375
                },
                {
                    "start": 684.13,
                    "end": 684.39,
                    "word": " about",
                    "probability": 0.8251953125
                },
                {
                    "start": 684.39,
                    "end": 684.55,
                    "word": " the",
                    "probability": 0.80859375
                },
                {
                    "start": 684.55,
                    "end": 684.95,
                    "word": " reports",
                    "probability": 0.70166015625
                },
                {
                    "start": 684.95,
                    "end": 685.23,
                    "word": ".",
                    "probability": 0.9755859375
                }
            ]
        },
        {
            "id": 88,
            "seek": 71319,
            "start": 685.23,
            "end": 689.27,
            "text": " The reports are the Mobility Reports.",
            "tokens": [
                50364,
                440,
                7122,
                366,
                264,
                37920,
                1140,
                45910,
                13,
                50514
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2865853716687458,
            "compression_ratio": 1.6265560165975104,
            "no_speech_prob": 0.010986328125,
            "words": [
                {
                    "start": 685.23,
                    "end": 685.41,
                    "word": " The",
                    "probability": 0.41748046875
                },
                {
                    "start": 685.41,
                    "end": 685.79,
                    "word": " reports",
                    "probability": 0.53955078125
                },
                {
                    "start": 685.79,
                    "end": 685.97,
                    "word": " are",
                    "probability": 0.06732177734375
                },
                {
                    "start": 685.97,
                    "end": 686.37,
                    "word": " the",
                    "probability": 0.68701171875
                },
                {
                    "start": 686.37,
                    "end": 688.39,
                    "word": " Mobility",
                    "probability": 0.537261962890625
                },
                {
                    "start": 688.39,
                    "end": 689.01,
                    "word": " Reports",
                    "probability": 0.52490234375
                },
                {
                    "start": 689.01,
                    "end": 689.27,
                    "word": ".",
                    "probability": 0.85107421875
                }
            ]
        },
        {
            "id": 89,
            "seek": 71319,
            "start": 689.27,
            "end": 695.63,
            "text": " These reports were the first product that Google launched that was built with differential privacy from the ground up.",
            "tokens": [
                50564,
                1981,
                7122,
                645,
                264,
                700,
                1674,
                300,
                3329,
                8730,
                300,
                390,
                3094,
                365,
                15756,
                11427,
                490,
                264,
                2727,
                493,
                13,
                50864
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2865853716687458,
            "compression_ratio": 1.6265560165975104,
            "no_speech_prob": 0.010986328125,
            "words": [
                {
                    "start": 689.27,
                    "end": 689.45,
                    "word": " These",
                    "probability": 0.88037109375
                },
                {
                    "start": 689.45,
                    "end": 689.83,
                    "word": " reports",
                    "probability": 0.70361328125
                },
                {
                    "start": 689.83,
                    "end": 690.03,
                    "word": " were",
                    "probability": 0.85107421875
                },
                {
                    "start": 690.03,
                    "end": 690.21,
                    "word": " the",
                    "probability": 0.8134765625
                },
                {
                    "start": 690.21,
                    "end": 690.47,
                    "word": " first",
                    "probability": 0.81298828125
                },
                {
                    "start": 690.47,
                    "end": 690.93,
                    "word": " product",
                    "probability": 0.67431640625
                },
                {
                    "start": 690.93,
                    "end": 691.09,
                    "word": " that",
                    "probability": 0.7763671875
                },
                {
                    "start": 691.09,
                    "end": 691.27,
                    "word": " Google",
                    "probability": 0.84228515625
                },
                {
                    "start": 691.27,
                    "end": 691.71,
                    "word": " launched",
                    "probability": 0.84375
                },
                {
                    "start": 691.71,
                    "end": 692.11,
                    "word": " that",
                    "probability": 0.426025390625
                },
                {
                    "start": 692.11,
                    "end": 692.31,
                    "word": " was",
                    "probability": 0.892578125
                },
                {
                    "start": 692.31,
                    "end": 692.63,
                    "word": " built",
                    "probability": 0.921875
                },
                {
                    "start": 692.63,
                    "end": 692.89,
                    "word": " with",
                    "probability": 0.81298828125
                },
                {
                    "start": 692.89,
                    "end": 693.33,
                    "word": " differential",
                    "probability": 0.64501953125
                },
                {
                    "start": 693.33,
                    "end": 693.79,
                    "word": " privacy",
                    "probability": 0.68798828125
                },
                {
                    "start": 693.79,
                    "end": 694.03,
                    "word": " from",
                    "probability": 0.82177734375
                },
                {
                    "start": 694.03,
                    "end": 694.17,
                    "word": " the",
                    "probability": 0.80859375
                },
                {
                    "start": 694.17,
                    "end": 694.47,
                    "word": " ground",
                    "probability": 0.78564453125
                },
                {
                    "start": 694.47,
                    "end": 694.91,
                    "word": " up",
                    "probability": 0.8974609375
                },
                {
                    "start": 694.91,
                    "end": 695.63,
                    "word": ".",
                    "probability": 0.9619140625
                }
            ]
        },
        {
            "id": 90,
            "seek": 71319,
            "start": 696.13,
            "end": 704.57,
            "text": " The reports do not share data about any individual's locations, movements, or conflicts.",
            "tokens": [
                50914,
                440,
                7122,
                360,
                406,
                2073,
                1412,
                466,
                604,
                2609,
                311,
                9253,
                11,
                9981,
                11,
                420,
                19807,
                13,
                51264
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2865853716687458,
            "compression_ratio": 1.6265560165975104,
            "no_speech_prob": 0.010986328125,
            "words": [
                {
                    "start": 696.13,
                    "end": 696.85,
                    "word": " The",
                    "probability": 0.802734375
                },
                {
                    "start": 696.85,
                    "end": 698.41,
                    "word": " reports",
                    "probability": 0.49462890625
                },
                {
                    "start": 698.41,
                    "end": 698.79,
                    "word": " do",
                    "probability": 0.328125
                },
                {
                    "start": 698.79,
                    "end": 698.93,
                    "word": " not",
                    "probability": 0.89697265625
                },
                {
                    "start": 698.93,
                    "end": 699.21,
                    "word": " share",
                    "probability": 0.8125
                },
                {
                    "start": 699.21,
                    "end": 699.97,
                    "word": " data",
                    "probability": 0.75244140625
                },
                {
                    "start": 699.97,
                    "end": 700.59,
                    "word": " about",
                    "probability": 0.81787109375
                },
                {
                    "start": 700.59,
                    "end": 701.13,
                    "word": " any",
                    "probability": 0.818359375
                },
                {
                    "start": 701.13,
                    "end": 701.67,
                    "word": " individual's",
                    "probability": 0.779052734375
                },
                {
                    "start": 701.67,
                    "end": 702.07,
                    "word": " locations",
                    "probability": 0.69189453125
                },
                {
                    "start": 702.07,
                    "end": 702.41,
                    "word": ",",
                    "probability": 0.9072265625
                },
                {
                    "start": 702.41,
                    "end": 702.85,
                    "word": " movements",
                    "probability": 0.84228515625
                },
                {
                    "start": 702.85,
                    "end": 703.01,
                    "word": ",",
                    "probability": 0.61376953125
                },
                {
                    "start": 703.01,
                    "end": 703.19,
                    "word": " or",
                    "probability": 0.9111328125
                },
                {
                    "start": 703.19,
                    "end": 703.85,
                    "word": " conflicts",
                    "probability": 0.399169921875
                },
                {
                    "start": 703.85,
                    "end": 704.57,
                    "word": ".",
                    "probability": 0.9873046875
                }
            ]
        },
        {
            "id": 91,
            "seek": 71319,
            "start": 704.95,
            "end": 714.93,
            "text": " The insights are created with aggregated, anonymized datasets from users who have turned on the location history setting, which is off by default.",
            "tokens": [
                51314,
                440,
                14310,
                366,
                2942,
                365,
                16743,
                770,
                11,
                37293,
                1602,
                42856,
                490,
                5022,
                567,
                362,
                3574,
                322,
                264,
                4914,
                2503,
                3287,
                11,
                597,
                307,
                766,
                538,
                7576,
                13,
                51814
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2865853716687458,
            "compression_ratio": 1.6265560165975104,
            "no_speech_prob": 0.010986328125,
            "words": [
                {
                    "start": 704.95,
                    "end": 705.09,
                    "word": " The",
                    "probability": 0.71435546875
                },
                {
                    "start": 705.09,
                    "end": 705.31,
                    "word": " insights",
                    "probability": 0.8525390625
                },
                {
                    "start": 705.31,
                    "end": 705.65,
                    "word": " are",
                    "probability": 0.89111328125
                },
                {
                    "start": 705.65,
                    "end": 706.09,
                    "word": " created",
                    "probability": 0.73876953125
                },
                {
                    "start": 706.09,
                    "end": 706.47,
                    "word": " with",
                    "probability": 0.810546875
                },
                {
                    "start": 706.47,
                    "end": 707.21,
                    "word": " aggregated",
                    "probability": 0.835693359375
                },
                {
                    "start": 707.21,
                    "end": 707.35,
                    "word": ",",
                    "probability": 0.56494140625
                },
                {
                    "start": 707.35,
                    "end": 708.05,
                    "word": " anonymized",
                    "probability": 0.84765625
                },
                {
                    "start": 708.05,
                    "end": 709.05,
                    "word": " datasets",
                    "probability": 0.42919921875
                },
                {
                    "start": 709.05,
                    "end": 709.43,
                    "word": " from",
                    "probability": 0.8046875
                },
                {
                    "start": 709.43,
                    "end": 709.77,
                    "word": " users",
                    "probability": 0.859375
                },
                {
                    "start": 709.77,
                    "end": 709.97,
                    "word": " who",
                    "probability": 0.80517578125
                },
                {
                    "start": 709.97,
                    "end": 710.23,
                    "word": " have",
                    "probability": 0.869140625
                },
                {
                    "start": 710.23,
                    "end": 710.51,
                    "word": " turned",
                    "probability": 0.75146484375
                },
                {
                    "start": 710.51,
                    "end": 710.97,
                    "word": " on",
                    "probability": 0.88525390625
                },
                {
                    "start": 710.97,
                    "end": 711.45,
                    "word": " the",
                    "probability": 0.70068359375
                },
                {
                    "start": 711.45,
                    "end": 711.79,
                    "word": " location",
                    "probability": 0.404052734375
                },
                {
                    "start": 711.79,
                    "end": 712.15,
                    "word": " history",
                    "probability": 0.73681640625
                },
                {
                    "start": 712.15,
                    "end": 712.73,
                    "word": " setting",
                    "probability": 0.77099609375
                },
                {
                    "start": 712.73,
                    "end": 713.07,
                    "word": ",",
                    "probability": 0.86865234375
                },
                {
                    "start": 713.07,
                    "end": 713.25,
                    "word": " which",
                    "probability": 0.9228515625
                },
                {
                    "start": 713.25,
                    "end": 713.43,
                    "word": " is",
                    "probability": 0.90673828125
                },
                {
                    "start": 713.43,
                    "end": 713.71,
                    "word": " off",
                    "probability": 0.80224609375
                },
                {
                    "start": 713.71,
                    "end": 713.93,
                    "word": " by",
                    "probability": 0.89599609375
                },
                {
                    "start": 714.35,
                    "end": 714.93,
                    "word": " default.",
                    "probability": 0.92919921875
                }
            ]
        },
        {
            "id": 92,
            "seek": 73989,
            "start": 715.23,
            "end": 719.59,
            "text": " People can turn off that setting at any time and delete their history.",
            "tokens": [
                50364,
                3432,
                393,
                1261,
                766,
                300,
                3287,
                412,
                604,
                565,
                293,
                12097,
                641,
                2503,
                13,
                50564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18677326309126477,
            "compression_ratio": 1.6016597510373445,
            "no_speech_prob": 0.001926422119140625,
            "words": [
                {
                    "start": 715.23,
                    "end": 715.27,
                    "word": " People",
                    "probability": 0.06475830078125
                },
                {
                    "start": 715.27,
                    "end": 715.53,
                    "word": " can",
                    "probability": 0.89306640625
                },
                {
                    "start": 715.53,
                    "end": 715.81,
                    "word": " turn",
                    "probability": 0.8154296875
                },
                {
                    "start": 715.81,
                    "end": 716.13,
                    "word": " off",
                    "probability": 0.78515625
                },
                {
                    "start": 716.13,
                    "end": 716.37,
                    "word": " that",
                    "probability": 0.83740234375
                },
                {
                    "start": 716.37,
                    "end": 716.71,
                    "word": " setting",
                    "probability": 0.82568359375
                },
                {
                    "start": 716.71,
                    "end": 716.97,
                    "word": " at",
                    "probability": 0.88037109375
                },
                {
                    "start": 716.97,
                    "end": 717.25,
                    "word": " any",
                    "probability": 0.77880859375
                },
                {
                    "start": 717.25,
                    "end": 717.57,
                    "word": " time",
                    "probability": 0.8017578125
                },
                {
                    "start": 717.57,
                    "end": 717.85,
                    "word": " and",
                    "probability": 0.8447265625
                },
                {
                    "start": 717.85,
                    "end": 718.19,
                    "word": " delete",
                    "probability": 0.85009765625
                },
                {
                    "start": 718.19,
                    "end": 718.47,
                    "word": " their",
                    "probability": 0.88916015625
                },
                {
                    "start": 718.47,
                    "end": 718.95,
                    "word": " history",
                    "probability": 0.84716796875
                },
                {
                    "start": 718.95,
                    "end": 719.59,
                    "word": ".",
                    "probability": 0.93359375
                }
            ]
        },
        {
            "id": 93,
            "seek": 73989,
            "start": 720.13,
            "end": 727.31,
            "text": " We took the additional step of utilizing differential privacy to ensure that no individual can be identified.",
            "tokens": [
                50614,
                492,
                1890,
                264,
                4497,
                1823,
                295,
                26775,
                15756,
                11427,
                281,
                5586,
                300,
                572,
                2609,
                393,
                312,
                9234,
                13,
                50964
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18677326309126477,
            "compression_ratio": 1.6016597510373445,
            "no_speech_prob": 0.001926422119140625,
            "words": [
                {
                    "start": 720.13,
                    "end": 720.37,
                    "word": " We",
                    "probability": 0.875
                },
                {
                    "start": 720.37,
                    "end": 720.57,
                    "word": " took",
                    "probability": 0.88623046875
                },
                {
                    "start": 720.57,
                    "end": 720.75,
                    "word": " the",
                    "probability": 0.80810546875
                },
                {
                    "start": 720.75,
                    "end": 721.13,
                    "word": " additional",
                    "probability": 0.8798828125
                },
                {
                    "start": 721.13,
                    "end": 721.63,
                    "word": " step",
                    "probability": 0.8232421875
                },
                {
                    "start": 721.63,
                    "end": 721.85,
                    "word": " of",
                    "probability": 0.90625
                },
                {
                    "start": 721.85,
                    "end": 722.51,
                    "word": " utilizing",
                    "probability": 0.70263671875
                },
                {
                    "start": 722.51,
                    "end": 723.01,
                    "word": " differential",
                    "probability": 0.71435546875
                },
                {
                    "start": 723.01,
                    "end": 723.57,
                    "word": " privacy",
                    "probability": 0.8271484375
                },
                {
                    "start": 723.57,
                    "end": 724.17,
                    "word": " to",
                    "probability": 0.818359375
                },
                {
                    "start": 724.17,
                    "end": 724.43,
                    "word": " ensure",
                    "probability": 0.84375
                },
                {
                    "start": 724.43,
                    "end": 724.67,
                    "word": " that",
                    "probability": 0.87255859375
                },
                {
                    "start": 724.67,
                    "end": 724.97,
                    "word": " no",
                    "probability": 0.9052734375
                },
                {
                    "start": 724.97,
                    "end": 725.49,
                    "word": " individual",
                    "probability": 0.716796875
                },
                {
                    "start": 725.49,
                    "end": 725.81,
                    "word": " can",
                    "probability": 0.89599609375
                },
                {
                    "start": 725.81,
                    "end": 726.11,
                    "word": " be",
                    "probability": 0.90625
                },
                {
                    "start": 726.11,
                    "end": 726.67,
                    "word": " identified",
                    "probability": 0.91064453125
                },
                {
                    "start": 726.67,
                    "end": 727.31,
                    "word": ".",
                    "probability": 0.984375
                }
            ]
        },
        {
            "id": 94,
            "seek": 73989,
            "start": 727.87,
            "end": 730.99,
            "text": " So how exactly are we using differential privacy?",
            "tokens": [
                51014,
                407,
                577,
                2293,
                366,
                321,
                1228,
                15756,
                11427,
                30,
                51114
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18677326309126477,
            "compression_ratio": 1.6016597510373445,
            "no_speech_prob": 0.001926422119140625,
            "words": [
                {
                    "start": 727.87,
                    "end": 728.05,
                    "word": " So",
                    "probability": 0.65869140625
                },
                {
                    "start": 728.05,
                    "end": 728.33,
                    "word": " how",
                    "probability": 0.7021484375
                },
                {
                    "start": 728.33,
                    "end": 728.69,
                    "word": " exactly",
                    "probability": 0.76904296875
                },
                {
                    "start": 728.69,
                    "end": 728.85,
                    "word": " are",
                    "probability": 0.8955078125
                },
                {
                    "start": 728.85,
                    "end": 728.99,
                    "word": " we",
                    "probability": 0.896484375
                },
                {
                    "start": 728.99,
                    "end": 729.35,
                    "word": " using",
                    "probability": 0.8291015625
                },
                {
                    "start": 729.35,
                    "end": 729.85,
                    "word": " differential",
                    "probability": 0.67822265625
                },
                {
                    "start": 729.85,
                    "end": 730.35,
                    "word": " privacy",
                    "probability": 0.83203125
                },
                {
                    "start": 730.35,
                    "end": 730.99,
                    "word": "?",
                    "probability": 0.98291015625
                }
            ]
        },
        {
            "id": 95,
            "seek": 73989,
            "start": 731.03,
            "end": 737.31,
            "text": " First, we produce a count of people at certain locations within a geographical area.",
            "tokens": [
                51164,
                2386,
                11,
                321,
                5258,
                257,
                1207,
                295,
                561,
                412,
                1629,
                9253,
                1951,
                257,
                39872,
                1859,
                13,
                51414
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18677326309126477,
            "compression_ratio": 1.6016597510373445,
            "no_speech_prob": 0.001926422119140625,
            "words": [
                {
                    "start": 731.03,
                    "end": 731.65,
                    "word": " First",
                    "probability": 0.89013671875
                },
                {
                    "start": 731.65,
                    "end": 732.07,
                    "word": ",",
                    "probability": 0.91259765625
                },
                {
                    "start": 732.07,
                    "end": 732.21,
                    "word": " we",
                    "probability": 0.90087890625
                },
                {
                    "start": 732.21,
                    "end": 732.55,
                    "word": " produce",
                    "probability": 0.53857421875
                },
                {
                    "start": 732.75,
                    "end": 733.01,
                    "word": " a count",
                    "probability": 0.7900390625
                },
                {
                    "start": 733.01,
                    "end": 733.23,
                    "word": " of",
                    "probability": 0.9052734375
                },
                {
                    "start": 733.23,
                    "end": 733.63,
                    "word": " people",
                    "probability": 0.92333984375
                },
                {
                    "start": 733.63,
                    "end": 733.95,
                    "word": " at",
                    "probability": 0.8984375
                },
                {
                    "start": 733.95,
                    "end": 734.31,
                    "word": " certain",
                    "probability": 0.30859375
                },
                {
                    "start": 734.31,
                    "end": 734.91,
                    "word": " locations",
                    "probability": 0.880859375
                },
                {
                    "start": 734.91,
                    "end": 735.61,
                    "word": " within",
                    "probability": 0.74853515625
                },
                {
                    "start": 735.87,
                    "end": 736.19,
                    "word": " a geographical",
                    "probability": 0.74853515625
                },
                {
                    "start": 736.19,
                    "end": 736.71,
                    "word": " area",
                    "probability": 0.8173828125
                },
                {
                    "start": 736.71,
                    "end": 737.31,
                    "word": ".",
                    "probability": 0.98095703125
                }
            ]
        },
        {
            "id": 96,
            "seek": 73989,
            "start": 737.31,
            "end": 741.93,
            "text": " Let's say that we count the visitors to parks in Vegas on a given day.",
            "tokens": [
                51464,
                961,
                311,
                584,
                300,
                321,
                1207,
                264,
                14315,
                281,
                16213,
                294,
                15841,
                322,
                257,
                2212,
                786,
                13,
                51664
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18677326309126477,
            "compression_ratio": 1.6016597510373445,
            "no_speech_prob": 0.001926422119140625,
            "words": [
                {
                    "start": 737.31,
                    "end": 737.57,
                    "word": " Let's",
                    "probability": 0.94189453125
                },
                {
                    "start": 737.57,
                    "end": 737.73,
                    "word": " say",
                    "probability": 0.896484375
                },
                {
                    "start": 737.73,
                    "end": 737.89,
                    "word": " that",
                    "probability": 0.85791015625
                },
                {
                    "start": 737.89,
                    "end": 738.05,
                    "word": " we",
                    "probability": 0.9033203125
                },
                {
                    "start": 738.05,
                    "end": 738.57,
                    "word": " count",
                    "probability": 0.810546875
                },
                {
                    "start": 738.57,
                    "end": 738.91,
                    "word": " the",
                    "probability": 0.73876953125
                },
                {
                    "start": 738.91,
                    "end": 739.37,
                    "word": " visitors",
                    "probability": 0.79833984375
                },
                {
                    "start": 739.37,
                    "end": 739.65,
                    "word": " to",
                    "probability": 0.90625
                },
                {
                    "start": 739.65,
                    "end": 739.99,
                    "word": " parks",
                    "probability": 0.75146484375
                },
                {
                    "start": 739.99,
                    "end": 740.23,
                    "word": " in",
                    "probability": 0.890625
                },
                {
                    "start": 740.23,
                    "end": 740.47,
                    "word": " Vegas",
                    "probability": 0.87255859375
                },
                {
                    "start": 740.47,
                    "end": 740.71,
                    "word": " on",
                    "probability": 0.8623046875
                },
                {
                    "start": 740.83,
                    "end": 741.05,
                    "word": " a given",
                    "probability": 0.77880859375
                },
                {
                    "start": 741.05,
                    "end": 741.33,
                    "word": " day",
                    "probability": 0.75732421875
                },
                {
                    "start": 741.33,
                    "end": 741.93,
                    "word": ".",
                    "probability": 0.90966796875
                }
            ]
        },
        {
            "id": 97,
            "seek": 76823,
            "start": 741.93,
            "end": 746.59,
            "text": " That might give us a number of 8,238 people.",
            "tokens": [
                50364,
                663,
                1062,
                976,
                505,
                257,
                1230,
                295,
                1649,
                11,
                9356,
                23,
                561,
                13,
                50564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.15934066458062812,
            "compression_ratio": 1.5092592592592593,
            "no_speech_prob": 0.0006985664367675781,
            "words": [
                {
                    "start": 741.93,
                    "end": 742.25,
                    "word": " That",
                    "probability": 0.546875
                },
                {
                    "start": 742.25,
                    "end": 742.49,
                    "word": " might",
                    "probability": 0.78173828125
                },
                {
                    "start": 742.49,
                    "end": 742.69,
                    "word": " give",
                    "probability": 0.8212890625
                },
                {
                    "start": 742.69,
                    "end": 742.89,
                    "word": " us",
                    "probability": 0.90869140625
                },
                {
                    "start": 743.25,
                    "end": 743.55,
                    "word": " a number",
                    "probability": 0.8828125
                },
                {
                    "start": 743.55,
                    "end": 743.83,
                    "word": " of",
                    "probability": 0.9033203125
                },
                {
                    "start": 743.83,
                    "end": 744.09,
                    "word": " 8",
                    "probability": 0.87255859375
                },
                {
                    "start": 744.09,
                    "end": 745.61,
                    "word": ",238",
                    "probability": 0.8564453125
                },
                {
                    "start": 745.61,
                    "end": 745.99,
                    "word": " people",
                    "probability": 0.9267578125
                },
                {
                    "start": 745.99,
                    "end": 746.59,
                    "word": ".",
                    "probability": 0.9873046875
                }
            ]
        },
        {
            "id": 98,
            "seek": 76823,
            "start": 747.03,
            "end": 751.17,
            "text": " Secondly, we add what we call some random noise to that number.",
            "tokens": [
                50614,
                19483,
                11,
                321,
                909,
                437,
                321,
                818,
                512,
                4974,
                5658,
                281,
                300,
                1230,
                13,
                50764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.15934066458062812,
            "compression_ratio": 1.5092592592592593,
            "no_speech_prob": 0.0006985664367675781,
            "words": [
                {
                    "start": 747.03,
                    "end": 747.63,
                    "word": " Secondly",
                    "probability": 0.69775390625
                },
                {
                    "start": 747.63,
                    "end": 747.87,
                    "word": ",",
                    "probability": 0.95947265625
                },
                {
                    "start": 747.87,
                    "end": 748.17,
                    "word": " we",
                    "probability": 0.904296875
                },
                {
                    "start": 748.17,
                    "end": 748.41,
                    "word": " add",
                    "probability": 0.7021484375
                },
                {
                    "start": 748.41,
                    "end": 748.53,
                    "word": " what",
                    "probability": 0.87255859375
                },
                {
                    "start": 748.53,
                    "end": 748.69,
                    "word": " we",
                    "probability": 0.90185546875
                },
                {
                    "start": 748.69,
                    "end": 748.93,
                    "word": " call",
                    "probability": 0.7939453125
                },
                {
                    "start": 748.93,
                    "end": 749.19,
                    "word": " some",
                    "probability": 0.7666015625
                },
                {
                    "start": 749.19,
                    "end": 749.45,
                    "word": " random",
                    "probability": 0.72119140625
                },
                {
                    "start": 749.45,
                    "end": 749.83,
                    "word": " noise",
                    "probability": 0.8837890625
                },
                {
                    "start": 749.83,
                    "end": 750.01,
                    "word": " to",
                    "probability": 0.9072265625
                },
                {
                    "start": 750.01,
                    "end": 750.27,
                    "word": " that",
                    "probability": 0.8896484375
                },
                {
                    "start": 750.27,
                    "end": 750.57,
                    "word": " number",
                    "probability": 0.87060546875
                },
                {
                    "start": 750.57,
                    "end": 751.17,
                    "word": ".",
                    "probability": 0.98681640625
                }
            ]
        },
        {
            "id": 99,
            "seek": 76823,
            "start": 751.31,
            "end": 757.55,
            "text": " Let's assume, for example, that we added an artificial number of 79.",
            "tokens": [
                50814,
                961,
                311,
                6552,
                11,
                337,
                1365,
                11,
                300,
                321,
                3869,
                364,
                11677,
                1230,
                295,
                32803,
                13,
                51114
            ],
            "temperature": 0.0,
            "avg_logprob": -0.15934066458062812,
            "compression_ratio": 1.5092592592592593,
            "no_speech_prob": 0.0006985664367675781,
            "words": [
                {
                    "start": 751.31,
                    "end": 751.59,
                    "word": " Let's",
                    "probability": 0.94580078125
                },
                {
                    "start": 751.59,
                    "end": 751.85,
                    "word": " assume",
                    "probability": 0.826171875
                },
                {
                    "start": 751.85,
                    "end": 751.99,
                    "word": ",",
                    "probability": 0.407958984375
                },
                {
                    "start": 751.99,
                    "end": 752.01,
                    "word": " for",
                    "probability": 0.890625
                },
                {
                    "start": 752.01,
                    "end": 752.47,
                    "word": " example",
                    "probability": 0.650390625
                },
                {
                    "start": 752.47,
                    "end": 752.67,
                    "word": ",",
                    "probability": 0.9609375
                },
                {
                    "start": 752.67,
                    "end": 752.75,
                    "word": " that",
                    "probability": 0.853515625
                },
                {
                    "start": 752.75,
                    "end": 752.89,
                    "word": " we",
                    "probability": 0.89501953125
                },
                {
                    "start": 752.89,
                    "end": 753.27,
                    "word": " added",
                    "probability": 0.69580078125
                },
                {
                    "start": 753.27,
                    "end": 754.59,
                    "word": " an",
                    "probability": 0.75732421875
                },
                {
                    "start": 754.59,
                    "end": 755.67,
                    "word": " artificial",
                    "probability": 0.796875
                },
                {
                    "start": 755.67,
                    "end": 756.13,
                    "word": " number",
                    "probability": 0.88525390625
                },
                {
                    "start": 756.13,
                    "end": 756.37,
                    "word": " of",
                    "probability": 0.90283203125
                },
                {
                    "start": 756.37,
                    "end": 756.95,
                    "word": " 79",
                    "probability": 0.88037109375
                },
                {
                    "start": 756.95,
                    "end": 757.55,
                    "word": ".",
                    "probability": 0.97119140625
                }
            ]
        },
        {
            "id": 100,
            "seek": 76823,
            "start": 757.87,
            "end": 762.95,
            "text": " We add these two and get a new count of 8,317.",
            "tokens": [
                51164,
                492,
                909,
                613,
                732,
                293,
                483,
                257,
                777,
                1207,
                295,
                1649,
                11,
                18,
                7773,
                13,
                51364
            ],
            "temperature": 0.0,
            "avg_logprob": -0.15934066458062812,
            "compression_ratio": 1.5092592592592593,
            "no_speech_prob": 0.0006985664367675781,
            "words": [
                {
                    "start": 757.87,
                    "end": 758.35,
                    "word": " We",
                    "probability": 0.90966796875
                },
                {
                    "start": 758.35,
                    "end": 758.73,
                    "word": " add",
                    "probability": 0.7998046875
                },
                {
                    "start": 758.73,
                    "end": 759.05,
                    "word": " these",
                    "probability": 0.61376953125
                },
                {
                    "start": 759.05,
                    "end": 759.35,
                    "word": " two",
                    "probability": 0.783203125
                },
                {
                    "start": 759.35,
                    "end": 759.65,
                    "word": " and",
                    "probability": 0.80810546875
                },
                {
                    "start": 759.65,
                    "end": 759.85,
                    "word": " get",
                    "probability": 0.88671875
                },
                {
                    "start": 760.05,
                    "end": 760.23,
                    "word": " a new",
                    "probability": 0.8154296875
                },
                {
                    "start": 760.23,
                    "end": 760.57,
                    "word": " count",
                    "probability": 0.80322265625
                },
                {
                    "start": 760.57,
                    "end": 760.75,
                    "word": " of",
                    "probability": 0.90625
                },
                {
                    "start": 760.75,
                    "end": 761.01,
                    "word": " 8",
                    "probability": 0.958984375
                },
                {
                    "start": 761.01,
                    "end": 762.35,
                    "word": ",317",
                    "probability": 0.96337890625
                },
                {
                    "start": 762.35,
                    "end": 762.95,
                    "word": ".",
                    "probability": 0.9912109375
                }
            ]
        },
        {
            "id": 101,
            "seek": 76823,
            "start": 763.31,
            "end": 770.27,
            "text": " This count is differentially private, and we can now use it to calculate the metrics that we publish.",
            "tokens": [
                51414,
                639,
                1207,
                307,
                819,
                2270,
                4551,
                11,
                293,
                321,
                393,
                586,
                764,
                309,
                281,
                8873,
                264,
                16367,
                300,
                321,
                11374,
                13,
                51764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.15934066458062812,
            "compression_ratio": 1.5092592592592593,
            "no_speech_prob": 0.0006985664367675781,
            "words": [
                {
                    "start": 763.31,
                    "end": 763.67,
                    "word": " This",
                    "probability": 0.7998046875
                },
                {
                    "start": 763.67,
                    "end": 764.07,
                    "word": " count",
                    "probability": 0.77001953125
                },
                {
                    "start": 764.07,
                    "end": 764.49,
                    "word": " is",
                    "probability": 0.90087890625
                },
                {
                    "start": 764.49,
                    "end": 765.41,
                    "word": " differentially",
                    "probability": 0.850341796875
                },
                {
                    "start": 765.41,
                    "end": 765.91,
                    "word": " private",
                    "probability": 0.79052734375
                },
                {
                    "start": 765.91,
                    "end": 766.21,
                    "word": ",",
                    "probability": 0.51025390625
                },
                {
                    "start": 766.21,
                    "end": 766.53,
                    "word": " and",
                    "probability": 0.89697265625
                },
                {
                    "start": 766.53,
                    "end": 766.67,
                    "word": " we",
                    "probability": 0.9033203125
                },
                {
                    "start": 766.67,
                    "end": 766.83,
                    "word": " can",
                    "probability": 0.9013671875
                },
                {
                    "start": 766.83,
                    "end": 767.21,
                    "word": " now",
                    "probability": 0.89306640625
                },
                {
                    "start": 767.21,
                    "end": 767.57,
                    "word": " use",
                    "probability": 0.81494140625
                },
                {
                    "start": 767.57,
                    "end": 767.71,
                    "word": " it",
                    "probability": 0.9033203125
                },
                {
                    "start": 767.71,
                    "end": 767.93,
                    "word": " to",
                    "probability": 0.90576171875
                },
                {
                    "start": 767.93,
                    "end": 768.27,
                    "word": " calculate",
                    "probability": 0.84423828125
                },
                {
                    "start": 768.27,
                    "end": 768.55,
                    "word": " the",
                    "probability": 0.80810546875
                },
                {
                    "start": 768.55,
                    "end": 768.95,
                    "word": " metrics",
                    "probability": 0.57958984375
                },
                {
                    "start": 768.95,
                    "end": 769.15,
                    "word": " that",
                    "probability": 0.8740234375
                },
                {
                    "start": 769.15,
                    "end": 769.33,
                    "word": " we",
                    "probability": 0.90673828125
                },
                {
                    "start": 769.33,
                    "end": 769.83,
                    "word": " publish",
                    "probability": 0.59716796875
                },
                {
                    "start": 769.83,
                    "end": 770.27,
                    "word": ".",
                    "probability": 0.9736328125
                }
            ]
        },
        {
            "id": 102,
            "seek": 79161,
            "start": 771.15,
            "end": 779.17,
            "text": " This is an example of a differentially private query that we do to produce reports like the mobility report.",
            "tokens": [
                50414,
                639,
                307,
                364,
                1365,
                295,
                257,
                819,
                2270,
                4551,
                14581,
                300,
                321,
                360,
                281,
                5258,
                7122,
                411,
                264,
                16199,
                2275,
                13,
                50764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1912841834127903,
            "compression_ratio": 1.4653465346534653,
            "no_speech_prob": 0.0011882781982421875,
            "words": [
                {
                    "start": 771.15,
                    "end": 771.99,
                    "word": " This",
                    "probability": 0.76318359375
                },
                {
                    "start": 771.99,
                    "end": 772.15,
                    "word": " is",
                    "probability": 0.90673828125
                },
                {
                    "start": 772.15,
                    "end": 772.29,
                    "word": " an",
                    "probability": 0.908203125
                },
                {
                    "start": 772.29,
                    "end": 772.63,
                    "word": " example",
                    "probability": 0.931640625
                },
                {
                    "start": 772.63,
                    "end": 773.13,
                    "word": " of",
                    "probability": 0.9013671875
                },
                {
                    "start": 773.37,
                    "end": 773.81,
                    "word": " a differentially",
                    "probability": 0.774658203125
                },
                {
                    "start": 773.81,
                    "end": 774.17,
                    "word": " private",
                    "probability": 0.2900390625
                },
                {
                    "start": 774.17,
                    "end": 774.63,
                    "word": " query",
                    "probability": 0.880859375
                },
                {
                    "start": 774.63,
                    "end": 774.85,
                    "word": " that",
                    "probability": 0.873046875
                },
                {
                    "start": 774.85,
                    "end": 775.05,
                    "word": " we",
                    "probability": 0.9111328125
                },
                {
                    "start": 775.05,
                    "end": 775.35,
                    "word": " do",
                    "probability": 0.89453125
                },
                {
                    "start": 775.35,
                    "end": 775.93,
                    "word": " to",
                    "probability": 0.88818359375
                },
                {
                    "start": 775.93,
                    "end": 776.37,
                    "word": " produce",
                    "probability": 0.72900390625
                },
                {
                    "start": 776.37,
                    "end": 776.97,
                    "word": " reports",
                    "probability": 0.76708984375
                },
                {
                    "start": 776.97,
                    "end": 777.23,
                    "word": " like",
                    "probability": 0.8544921875
                },
                {
                    "start": 777.23,
                    "end": 777.41,
                    "word": " the",
                    "probability": 0.80859375
                },
                {
                    "start": 777.41,
                    "end": 777.71,
                    "word": " mobility",
                    "probability": 0.6669921875
                },
                {
                    "start": 777.71,
                    "end": 778.33,
                    "word": " report",
                    "probability": 0.77099609375
                },
                {
                    "start": 778.33,
                    "end": 779.17,
                    "word": ".",
                    "probability": 0.97802734375
                }
            ]
        },
        {
            "id": 103,
            "seek": 79161,
            "start": 779.31,
            "end": 790.21,
            "text": " It's helpful for people who are familiar with the SQL environment because it uses the general syntax by only introducing new anonymization functions.",
            "tokens": [
                50814,
                467,
                311,
                4961,
                337,
                561,
                567,
                366,
                4963,
                365,
                264,
                19200,
                2823,
                570,
                309,
                4960,
                264,
                2674,
                28431,
                538,
                787,
                15424,
                777,
                37293,
                2144,
                6828,
                13,
                51314
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1912841834127903,
            "compression_ratio": 1.4653465346534653,
            "no_speech_prob": 0.0011882781982421875,
            "words": [
                {
                    "start": 779.31,
                    "end": 779.99,
                    "word": " It's",
                    "probability": 0.907958984375
                },
                {
                    "start": 779.99,
                    "end": 780.43,
                    "word": " helpful",
                    "probability": 0.82080078125
                },
                {
                    "start": 780.43,
                    "end": 780.79,
                    "word": " for",
                    "probability": 0.89306640625
                },
                {
                    "start": 780.79,
                    "end": 781.05,
                    "word": " people",
                    "probability": 0.921875
                },
                {
                    "start": 781.05,
                    "end": 781.29,
                    "word": " who",
                    "probability": 0.78564453125
                },
                {
                    "start": 781.29,
                    "end": 781.41,
                    "word": " are",
                    "probability": 0.900390625
                },
                {
                    "start": 781.41,
                    "end": 781.73,
                    "word": " familiar",
                    "probability": 0.7744140625
                },
                {
                    "start": 781.73,
                    "end": 782.13,
                    "word": " with",
                    "probability": 0.8193359375
                },
                {
                    "start": 782.13,
                    "end": 782.31,
                    "word": " the",
                    "probability": 0.7705078125
                },
                {
                    "start": 782.31,
                    "end": 782.51,
                    "word": " SQL",
                    "probability": 0.89990234375
                },
                {
                    "start": 782.51,
                    "end": 783.15,
                    "word": " environment",
                    "probability": 0.71630859375
                },
                {
                    "start": 783.15,
                    "end": 783.89,
                    "word": " because",
                    "probability": 0.59716796875
                },
                {
                    "start": 783.89,
                    "end": 784.31,
                    "word": " it",
                    "probability": 0.9033203125
                },
                {
                    "start": 784.31,
                    "end": 784.85,
                    "word": " uses",
                    "probability": 0.80126953125
                },
                {
                    "start": 784.85,
                    "end": 785.27,
                    "word": " the",
                    "probability": 0.8046875
                },
                {
                    "start": 785.27,
                    "end": 785.69,
                    "word": " general",
                    "probability": 0.73876953125
                },
                {
                    "start": 785.69,
                    "end": 786.23,
                    "word": " syntax",
                    "probability": 0.90966796875
                },
                {
                    "start": 786.23,
                    "end": 786.73,
                    "word": " by",
                    "probability": 0.8837890625
                },
                {
                    "start": 786.73,
                    "end": 787.07,
                    "word": " only",
                    "probability": 0.84423828125
                },
                {
                    "start": 787.07,
                    "end": 787.61,
                    "word": " introducing",
                    "probability": 0.71728515625
                },
                {
                    "start": 787.61,
                    "end": 788.09,
                    "word": " new",
                    "probability": 0.75537109375
                },
                {
                    "start": 788.09,
                    "end": 789.07,
                    "word": " anonymization",
                    "probability": 0.8271484375
                },
                {
                    "start": 789.07,
                    "end": 789.73,
                    "word": " functions",
                    "probability": 0.8623046875
                },
                {
                    "start": 789.73,
                    "end": 790.21,
                    "word": ".",
                    "probability": 0.5576171875
                }
            ]
        },
        {
            "id": 104,
            "seek": 79161,
            "start": 790.21,
            "end": 793.65,
            "text": " In this case, an anonymization count.",
            "tokens": [
                51364,
                682,
                341,
                1389,
                11,
                364,
                37293,
                2144,
                1207,
                13,
                51514
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1912841834127903,
            "compression_ratio": 1.4653465346534653,
            "no_speech_prob": 0.0011882781982421875,
            "words": [
                {
                    "start": 790.21,
                    "end": 790.39,
                    "word": " In",
                    "probability": 0.904296875
                },
                {
                    "start": 790.39,
                    "end": 790.59,
                    "word": " this",
                    "probability": 0.88037109375
                },
                {
                    "start": 790.59,
                    "end": 790.97,
                    "word": " case",
                    "probability": 0.80908203125
                },
                {
                    "start": 790.97,
                    "end": 791.59,
                    "word": ",",
                    "probability": 0.9326171875
                },
                {
                    "start": 791.59,
                    "end": 791.73,
                    "word": " an",
                    "probability": 0.483154296875
                },
                {
                    "start": 791.73,
                    "end": 792.71,
                    "word": " anonymization",
                    "probability": 0.871826171875
                },
                {
                    "start": 792.71,
                    "end": 793.23,
                    "word": " count",
                    "probability": 0.74267578125
                },
                {
                    "start": 793.23,
                    "end": 793.65,
                    "word": ".",
                    "probability": 0.9921875
                }
            ]
        },
        {
            "id": 105,
            "seek": 81487,
            "start": 794.91,
            "end": 806.11,
            "text": " Generally, to produce the metrics, in the reports, we compute a baseline using differential privacy for a given weekday in the past.",
            "tokens": [
                50414,
                21082,
                11,
                281,
                5258,
                264,
                16367,
                11,
                294,
                264,
                7122,
                11,
                321,
                14722,
                257,
                20518,
                1228,
                15756,
                11427,
                337,
                257,
                2212,
                1243,
                810,
                294,
                264,
                1791,
                13,
                50914
            ],
            "temperature": 0.0,
            "avg_logprob": -0.21187034293786802,
            "compression_ratio": 1.4943820224719102,
            "no_speech_prob": 0.04144287109375,
            "words": [
                {
                    "start": 794.91,
                    "end": 795.53,
                    "word": " Generally",
                    "probability": 0.5966796875
                },
                {
                    "start": 795.53,
                    "end": 795.69,
                    "word": ",",
                    "probability": 0.436279296875
                },
                {
                    "start": 795.69,
                    "end": 795.79,
                    "word": " to",
                    "probability": 0.9033203125
                },
                {
                    "start": 795.79,
                    "end": 796.09,
                    "word": " produce",
                    "probability": 0.7265625
                },
                {
                    "start": 796.09,
                    "end": 796.27,
                    "word": " the",
                    "probability": 0.80224609375
                },
                {
                    "start": 796.27,
                    "end": 796.65,
                    "word": " metrics",
                    "probability": 0.73583984375
                },
                {
                    "start": 796.65,
                    "end": 797.93,
                    "word": ",",
                    "probability": 0.67822265625
                },
                {
                    "start": 797.93,
                    "end": 799.39,
                    "word": " in",
                    "probability": 0.285888671875
                },
                {
                    "start": 799.39,
                    "end": 799.69,
                    "word": " the",
                    "probability": 0.78515625
                },
                {
                    "start": 799.69,
                    "end": 800.09,
                    "word": " reports",
                    "probability": 0.70263671875
                },
                {
                    "start": 800.09,
                    "end": 800.21,
                    "word": ",",
                    "probability": 0.75
                },
                {
                    "start": 800.21,
                    "end": 800.31,
                    "word": " we",
                    "probability": 0.896484375
                },
                {
                    "start": 800.31,
                    "end": 800.79,
                    "word": " compute",
                    "probability": 0.87255859375
                },
                {
                    "start": 801.11,
                    "end": 801.55,
                    "word": " a baseline",
                    "probability": 0.86767578125
                },
                {
                    "start": 801.55,
                    "end": 802.69,
                    "word": " using",
                    "probability": 0.7529296875
                },
                {
                    "start": 802.69,
                    "end": 803.13,
                    "word": " differential",
                    "probability": 0.57958984375
                },
                {
                    "start": 803.13,
                    "end": 803.55,
                    "word": " privacy",
                    "probability": 0.8408203125
                },
                {
                    "start": 803.55,
                    "end": 803.89,
                    "word": " for",
                    "probability": 0.873046875
                },
                {
                    "start": 804.07,
                    "end": 804.21,
                    "word": " a given",
                    "probability": 0.78125
                },
                {
                    "start": 804.21,
                    "end": 804.73,
                    "word": " weekday",
                    "probability": 0.727294921875
                },
                {
                    "start": 804.73,
                    "end": 804.99,
                    "word": " in",
                    "probability": 0.89599609375
                },
                {
                    "start": 804.99,
                    "end": 805.11,
                    "word": " the",
                    "probability": 0.81396484375
                },
                {
                    "start": 805.11,
                    "end": 805.49,
                    "word": " past",
                    "probability": 0.80712890625
                },
                {
                    "start": 805.49,
                    "end": 806.11,
                    "word": ".",
                    "probability": 0.96533203125
                }
            ]
        },
        {
            "id": 106,
            "seek": 81487,
            "start": 806.31,
            "end": 810.33,
            "text": " Second, we compute those same metrics for a given day.",
            "tokens": [
                50964,
                5736,
                11,
                321,
                14722,
                729,
                912,
                16367,
                337,
                257,
                2212,
                786,
                13,
                51164
            ],
            "temperature": 0.0,
            "avg_logprob": -0.21187034293786802,
            "compression_ratio": 1.4943820224719102,
            "no_speech_prob": 0.04144287109375,
            "words": [
                {
                    "start": 806.31,
                    "end": 806.93,
                    "word": " Second",
                    "probability": 0.71240234375
                },
                {
                    "start": 806.93,
                    "end": 807.19,
                    "word": ",",
                    "probability": 0.99072265625
                },
                {
                    "start": 807.19,
                    "end": 807.33,
                    "word": " we",
                    "probability": 0.90380859375
                },
                {
                    "start": 807.33,
                    "end": 807.63,
                    "word": " compute",
                    "probability": 0.87646484375
                },
                {
                    "start": 807.63,
                    "end": 807.87,
                    "word": " those",
                    "probability": 0.73779296875
                },
                {
                    "start": 807.87,
                    "end": 808.95,
                    "word": " same",
                    "probability": 0.461181640625
                },
                {
                    "start": 808.95,
                    "end": 809.23,
                    "word": " metrics",
                    "probability": 0.82861328125
                },
                {
                    "start": 809.23,
                    "end": 809.47,
                    "word": " for",
                    "probability": 0.89453125
                },
                {
                    "start": 809.61,
                    "end": 809.79,
                    "word": " a given",
                    "probability": 0.78759765625
                },
                {
                    "start": 809.79,
                    "end": 810.07,
                    "word": " day",
                    "probability": 0.8916015625
                },
                {
                    "start": 810.07,
                    "end": 810.33,
                    "word": ".",
                    "probability": 0.73974609375
                }
            ]
        },
        {
            "id": 107,
            "seek": 81487,
            "start": 810.33,
            "end": 816.91,
            "text": " Let's say for Friday, March 27, 5,000 people visited parks in the same region.",
            "tokens": [
                51214,
                961,
                311,
                584,
                337,
                6984,
                11,
                6129,
                7634,
                11,
                1025,
                11,
                1360,
                561,
                11220,
                16213,
                294,
                264,
                912,
                4458,
                13,
                51464
            ],
            "temperature": 0.0,
            "avg_logprob": -0.21187034293786802,
            "compression_ratio": 1.4943820224719102,
            "no_speech_prob": 0.04144287109375,
            "words": [
                {
                    "start": 810.33,
                    "end": 810.55,
                    "word": " Let's",
                    "probability": 0.9326171875
                },
                {
                    "start": 810.55,
                    "end": 810.77,
                    "word": " say",
                    "probability": 0.89697265625
                },
                {
                    "start": 810.77,
                    "end": 811.11,
                    "word": " for",
                    "probability": 0.68896484375
                },
                {
                    "start": 811.11,
                    "end": 811.55,
                    "word": " Friday",
                    "probability": 0.8095703125
                },
                {
                    "start": 811.55,
                    "end": 811.67,
                    "word": ",",
                    "probability": 0.90625
                },
                {
                    "start": 811.67,
                    "end": 811.89,
                    "word": " March",
                    "probability": 0.9033203125
                },
                {
                    "start": 811.89,
                    "end": 812.41,
                    "word": " 27",
                    "probability": 0.87451171875
                },
                {
                    "start": 812.41,
                    "end": 813.07,
                    "word": ",",
                    "probability": 0.69677734375
                },
                {
                    "start": 813.07,
                    "end": 813.61,
                    "word": " 5",
                    "probability": 0.49169921875
                },
                {
                    "start": 813.61,
                    "end": 813.87,
                    "word": ",000",
                    "probability": 0.9912109375
                },
                {
                    "start": 813.87,
                    "end": 814.25,
                    "word": " people",
                    "probability": 0.92333984375
                },
                {
                    "start": 814.25,
                    "end": 814.71,
                    "word": " visited",
                    "probability": 0.8818359375
                },
                {
                    "start": 814.71,
                    "end": 815.15,
                    "word": " parks",
                    "probability": 0.60400390625
                },
                {
                    "start": 815.15,
                    "end": 815.47,
                    "word": " in",
                    "probability": 0.90478515625
                },
                {
                    "start": 815.47,
                    "end": 815.73,
                    "word": " the",
                    "probability": 0.810546875
                },
                {
                    "start": 815.73,
                    "end": 815.95,
                    "word": " same",
                    "probability": 0.81201171875
                },
                {
                    "start": 815.95,
                    "end": 816.29,
                    "word": " region",
                    "probability": 0.77587890625
                },
                {
                    "start": 816.29,
                    "end": 816.91,
                    "word": ".",
                    "probability": 0.9677734375
                }
            ]
        },
        {
            "id": 108,
            "seek": 84303,
            "start": 816.91,
            "end": 825.53,
            "text": " Then we divide that result by the first result, showing that there was a 50% reduction in visits to parks in that region.",
            "tokens": [
                50364,
                1396,
                321,
                9845,
                300,
                1874,
                538,
                264,
                700,
                1874,
                11,
                4099,
                300,
                456,
                390,
                257,
                2625,
                4,
                11004,
                294,
                17753,
                281,
                16213,
                294,
                300,
                4458,
                13,
                50764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20940722141069235,
            "compression_ratio": 1.7419354838709677,
            "no_speech_prob": 0.00884246826171875,
            "words": [
                {
                    "start": 816.91,
                    "end": 817.39,
                    "word": " Then",
                    "probability": 0.64892578125
                },
                {
                    "start": 817.39,
                    "end": 817.63,
                    "word": " we",
                    "probability": 0.779296875
                },
                {
                    "start": 817.63,
                    "end": 818.03,
                    "word": " divide",
                    "probability": 0.77978515625
                },
                {
                    "start": 818.03,
                    "end": 818.91,
                    "word": " that",
                    "probability": 0.654296875
                },
                {
                    "start": 818.91,
                    "end": 819.33,
                    "word": " result",
                    "probability": 0.74072265625
                },
                {
                    "start": 819.33,
                    "end": 819.55,
                    "word": " by",
                    "probability": 0.89697265625
                },
                {
                    "start": 819.55,
                    "end": 819.75,
                    "word": " the",
                    "probability": 0.81494140625
                },
                {
                    "start": 819.75,
                    "end": 820.01,
                    "word": " first",
                    "probability": 0.8046875
                },
                {
                    "start": 820.01,
                    "end": 820.47,
                    "word": " result",
                    "probability": 0.5380859375
                },
                {
                    "start": 820.47,
                    "end": 820.83,
                    "word": ",",
                    "probability": 0.5341796875
                },
                {
                    "start": 820.83,
                    "end": 821.13,
                    "word": " showing",
                    "probability": 0.77490234375
                },
                {
                    "start": 821.13,
                    "end": 821.29,
                    "word": " that",
                    "probability": 0.83642578125
                },
                {
                    "start": 821.29,
                    "end": 821.39,
                    "word": " there",
                    "probability": 0.84130859375
                },
                {
                    "start": 821.39,
                    "end": 821.53,
                    "word": " was",
                    "probability": 0.890625
                },
                {
                    "start": 821.65,
                    "end": 821.95,
                    "word": " a 50",
                    "probability": 0.87255859375
                },
                {
                    "start": 821.95,
                    "end": 822.19,
                    "word": "%",
                    "probability": 0.90234375
                },
                {
                    "start": 822.19,
                    "end": 822.65,
                    "word": " reduction",
                    "probability": 0.86474609375
                },
                {
                    "start": 822.65,
                    "end": 823.23,
                    "word": " in",
                    "probability": 0.66796875
                },
                {
                    "start": 823.23,
                    "end": 823.69,
                    "word": " visits",
                    "probability": 0.9482421875
                },
                {
                    "start": 823.69,
                    "end": 823.91,
                    "word": " to",
                    "probability": 0.89990234375
                },
                {
                    "start": 823.91,
                    "end": 824.23,
                    "word": " parks",
                    "probability": 0.7197265625
                },
                {
                    "start": 824.23,
                    "end": 824.47,
                    "word": " in",
                    "probability": 0.90283203125
                },
                {
                    "start": 824.47,
                    "end": 824.69,
                    "word": " that",
                    "probability": 0.88134765625
                },
                {
                    "start": 824.69,
                    "end": 825.03,
                    "word": " region",
                    "probability": 0.76220703125
                },
                {
                    "start": 825.03,
                    "end": 825.53,
                    "word": ".",
                    "probability": 0.82275390625
                }
            ]
        },
        {
            "id": 109,
            "seek": 84303,
            "start": 825.53,
            "end": 829.21,
            "text": " And then we repeat this process for all the metrics that we are producing.",
            "tokens": [
                50814,
                400,
                550,
                321,
                7149,
                341,
                1399,
                337,
                439,
                264,
                16367,
                300,
                321,
                366,
                10501,
                13,
                50964
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20940722141069235,
            "compression_ratio": 1.7419354838709677,
            "no_speech_prob": 0.00884246826171875,
            "words": [
                {
                    "start": 825.53,
                    "end": 825.71,
                    "word": " And",
                    "probability": 0.7666015625
                },
                {
                    "start": 825.71,
                    "end": 825.87,
                    "word": " then",
                    "probability": 0.7109375
                },
                {
                    "start": 825.87,
                    "end": 826.05,
                    "word": " we",
                    "probability": 0.87548828125
                },
                {
                    "start": 826.05,
                    "end": 826.35,
                    "word": " repeat",
                    "probability": 0.71337890625
                },
                {
                    "start": 826.35,
                    "end": 826.55,
                    "word": " this",
                    "probability": 0.87255859375
                },
                {
                    "start": 826.55,
                    "end": 826.85,
                    "word": " process",
                    "probability": 0.8369140625
                },
                {
                    "start": 826.85,
                    "end": 827.07,
                    "word": " for",
                    "probability": 0.89501953125
                },
                {
                    "start": 827.07,
                    "end": 827.29,
                    "word": " all",
                    "probability": 0.9033203125
                },
                {
                    "start": 827.29,
                    "end": 827.49,
                    "word": " the",
                    "probability": 0.67431640625
                },
                {
                    "start": 827.49,
                    "end": 827.75,
                    "word": " metrics",
                    "probability": 0.806640625
                },
                {
                    "start": 827.75,
                    "end": 827.93,
                    "word": " that",
                    "probability": 0.86181640625
                },
                {
                    "start": 827.93,
                    "end": 828.07,
                    "word": " we",
                    "probability": 0.9033203125
                },
                {
                    "start": 828.07,
                    "end": 828.21,
                    "word": " are",
                    "probability": 0.8837890625
                },
                {
                    "start": 828.21,
                    "end": 828.65,
                    "word": " producing",
                    "probability": 0.7412109375
                },
                {
                    "start": 828.65,
                    "end": 829.21,
                    "word": ".",
                    "probability": 0.7578125
                }
            ]
        },
        {
            "id": 110,
            "seek": 84303,
            "start": 829.21,
            "end": 833.91,
            "text": " And therefore, all the metrics that we're reporting are produced with differential privacy.",
            "tokens": [
                51014,
                400,
                4412,
                11,
                439,
                264,
                16367,
                300,
                321,
                434,
                10031,
                366,
                7126,
                365,
                15756,
                11427,
                13,
                51214
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20940722141069235,
            "compression_ratio": 1.7419354838709677,
            "no_speech_prob": 0.00884246826171875,
            "words": [
                {
                    "start": 829.21,
                    "end": 829.35,
                    "word": " And",
                    "probability": 0.82275390625
                },
                {
                    "start": 829.35,
                    "end": 829.65,
                    "word": " therefore",
                    "probability": 0.697265625
                },
                {
                    "start": 829.65,
                    "end": 829.91,
                    "word": ",",
                    "probability": 0.53369140625
                },
                {
                    "start": 829.91,
                    "end": 829.99,
                    "word": " all",
                    "probability": 0.8955078125
                },
                {
                    "start": 829.99,
                    "end": 830.21,
                    "word": " the",
                    "probability": 0.4765625
                },
                {
                    "start": 830.21,
                    "end": 830.53,
                    "word": " metrics",
                    "probability": 0.78271484375
                },
                {
                    "start": 830.53,
                    "end": 830.73,
                    "word": " that",
                    "probability": 0.86279296875
                },
                {
                    "start": 830.73,
                    "end": 830.91,
                    "word": " we're",
                    "probability": 0.8369140625
                },
                {
                    "start": 830.91,
                    "end": 831.29,
                    "word": " reporting",
                    "probability": 0.7529296875
                },
                {
                    "start": 831.29,
                    "end": 831.71,
                    "word": " are",
                    "probability": 0.88525390625
                },
                {
                    "start": 831.71,
                    "end": 832.23,
                    "word": " produced",
                    "probability": 0.8056640625
                },
                {
                    "start": 832.23,
                    "end": 832.55,
                    "word": " with",
                    "probability": 0.81591796875
                },
                {
                    "start": 832.55,
                    "end": 832.89,
                    "word": " differential",
                    "probability": 0.6953125
                },
                {
                    "start": 832.89,
                    "end": 833.35,
                    "word": " privacy",
                    "probability": 0.489013671875
                },
                {
                    "start": 833.35,
                    "end": 833.91,
                    "word": ".",
                    "probability": 0.99072265625
                }
            ]
        },
        {
            "id": 111,
            "seek": 84303,
            "start": 834.73,
            "end": 840.41,
            "text": " There are some interesting practical aspects on how we use these reports.",
            "tokens": [
                51264,
                821,
                366,
                512,
                1880,
                8496,
                7270,
                322,
                577,
                321,
                764,
                613,
                7122,
                13,
                51514
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20940722141069235,
            "compression_ratio": 1.7419354838709677,
            "no_speech_prob": 0.00884246826171875,
            "words": [
                {
                    "start": 834.73,
                    "end": 835.29,
                    "word": " There",
                    "probability": 0.66162109375
                },
                {
                    "start": 835.29,
                    "end": 835.41,
                    "word": " are",
                    "probability": 0.61962890625
                },
                {
                    "start": 835.41,
                    "end": 835.65,
                    "word": " some",
                    "probability": 0.806640625
                },
                {
                    "start": 835.65,
                    "end": 836.13,
                    "word": " interesting",
                    "probability": 0.642578125
                },
                {
                    "start": 836.13,
                    "end": 836.61,
                    "word": " practical",
                    "probability": 0.7919921875
                },
                {
                    "start": 836.61,
                    "end": 837.13,
                    "word": " aspects",
                    "probability": 0.75927734375
                },
                {
                    "start": 837.13,
                    "end": 838.15,
                    "word": " on",
                    "probability": 0.86572265625
                },
                {
                    "start": 838.15,
                    "end": 838.39,
                    "word": " how",
                    "probability": 0.90283203125
                },
                {
                    "start": 838.39,
                    "end": 838.57,
                    "word": " we",
                    "probability": 0.87060546875
                },
                {
                    "start": 838.57,
                    "end": 839.25,
                    "word": " use",
                    "probability": 0.322265625
                },
                {
                    "start": 839.25,
                    "end": 839.49,
                    "word": " these",
                    "probability": 0.71337890625
                },
                {
                    "start": 839.49,
                    "end": 839.91,
                    "word": " reports",
                    "probability": 0.74609375
                },
                {
                    "start": 839.91,
                    "end": 840.41,
                    "word": ".",
                    "probability": 0.9794921875
                }
            ]
        },
        {
            "id": 112,
            "seek": 84303,
            "start": 840.41,
            "end": 845.07,
            "text": " We limit the number of times that a person is counted on each metric.",
            "tokens": [
                51564,
                492,
                4948,
                264,
                1230,
                295,
                1413,
                300,
                257,
                954,
                307,
                20150,
                322,
                1184,
                20678,
                13,
                51764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20940722141069235,
            "compression_ratio": 1.7419354838709677,
            "no_speech_prob": 0.00884246826171875,
            "words": [
                {
                    "start": 840.41,
                    "end": 840.67,
                    "word": " We",
                    "probability": 0.8115234375
                },
                {
                    "start": 840.67,
                    "end": 840.95,
                    "word": " limit",
                    "probability": 0.7294921875
                },
                {
                    "start": 840.95,
                    "end": 841.35,
                    "word": " the",
                    "probability": 0.810546875
                },
                {
                    "start": 841.35,
                    "end": 841.59,
                    "word": " number",
                    "probability": 0.8876953125
                },
                {
                    "start": 841.59,
                    "end": 841.81,
                    "word": " of",
                    "probability": 0.90625
                },
                {
                    "start": 841.81,
                    "end": 842.17,
                    "word": " times",
                    "probability": 0.8369140625
                },
                {
                    "start": 842.17,
                    "end": 842.65,
                    "word": " that",
                    "probability": 0.86328125
                },
                {
                    "start": 842.83,
                    "end": 843.19,
                    "word": " a person",
                    "probability": 0.7998046875
                },
                {
                    "start": 843.19,
                    "end": 843.47,
                    "word": " is",
                    "probability": 0.90283203125
                },
                {
                    "start": 843.47,
                    "end": 843.73,
                    "word": " counted",
                    "probability": 0.80029296875
                },
                {
                    "start": 843.73,
                    "end": 843.97,
                    "word": " on",
                    "probability": 0.900390625
                },
                {
                    "start": 843.97,
                    "end": 844.19,
                    "word": " each",
                    "probability": 0.8623046875
                },
                {
                    "start": 844.19,
                    "end": 844.51,
                    "word": " metric",
                    "probability": 0.8671875
                },
                {
                    "start": 844.51,
                    "end": 845.07,
                    "word": ".",
                    "probability": 0.97412109375
                }
            ]
        },
        {
            "id": 113,
            "seek": 87153,
            "start": 845.07,
            "end": 852.29,
            "text": " For example, if for some reason I go to the grocery store 100 times in a day, I will only be counted once.",
            "tokens": [
                50364,
                1171,
                1365,
                11,
                498,
                337,
                512,
                1778,
                286,
                352,
                281,
                264,
                14410,
                3531,
                2319,
                1413,
                294,
                257,
                786,
                11,
                286,
                486,
                787,
                312,
                20150,
                1564,
                13,
                50664
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1643350252578425,
            "compression_ratio": 1.4814814814814814,
            "no_speech_prob": 0.0025119781494140625,
            "words": [
                {
                    "start": 845.07,
                    "end": 845.57,
                    "word": " For",
                    "probability": 0.7451171875
                },
                {
                    "start": 845.57,
                    "end": 845.99,
                    "word": " example",
                    "probability": 0.93408203125
                },
                {
                    "start": 845.99,
                    "end": 846.37,
                    "word": ",",
                    "probability": 0.9130859375
                },
                {
                    "start": 846.37,
                    "end": 846.87,
                    "word": " if",
                    "probability": 0.87060546875
                },
                {
                    "start": 846.87,
                    "end": 847.05,
                    "word": " for",
                    "probability": 0.845703125
                },
                {
                    "start": 847.05,
                    "end": 847.29,
                    "word": " some",
                    "probability": 0.8125
                },
                {
                    "start": 847.29,
                    "end": 847.47,
                    "word": " reason",
                    "probability": 0.919921875
                },
                {
                    "start": 847.47,
                    "end": 847.67,
                    "word": " I",
                    "probability": 0.9248046875
                },
                {
                    "start": 847.67,
                    "end": 847.81,
                    "word": " go",
                    "probability": 0.904296875
                },
                {
                    "start": 847.81,
                    "end": 847.89,
                    "word": " to",
                    "probability": 0.89404296875
                },
                {
                    "start": 847.89,
                    "end": 848.05,
                    "word": " the",
                    "probability": 0.75146484375
                },
                {
                    "start": 848.05,
                    "end": 848.31,
                    "word": " grocery",
                    "probability": 0.6962890625
                },
                {
                    "start": 848.31,
                    "end": 848.75,
                    "word": " store",
                    "probability": 0.802734375
                },
                {
                    "start": 848.75,
                    "end": 849.25,
                    "word": " 100",
                    "probability": 0.335693359375
                },
                {
                    "start": 849.25,
                    "end": 849.65,
                    "word": " times",
                    "probability": 0.78564453125
                },
                {
                    "start": 849.65,
                    "end": 849.77,
                    "word": " in",
                    "probability": 0.8046875
                },
                {
                    "start": 849.91,
                    "end": 850.09,
                    "word": " a day",
                    "probability": 0.896484375
                },
                {
                    "start": 850.09,
                    "end": 850.23,
                    "word": ",",
                    "probability": 0.77734375
                },
                {
                    "start": 850.23,
                    "end": 850.33,
                    "word": " I",
                    "probability": 0.99072265625
                },
                {
                    "start": 850.33,
                    "end": 850.49,
                    "word": " will",
                    "probability": 0.77001953125
                },
                {
                    "start": 850.49,
                    "end": 850.77,
                    "word": " only",
                    "probability": 0.84033203125
                },
                {
                    "start": 850.77,
                    "end": 850.95,
                    "word": " be",
                    "probability": 0.873046875
                },
                {
                    "start": 850.95,
                    "end": 851.23,
                    "word": " counted",
                    "probability": 0.7822265625
                },
                {
                    "start": 851.23,
                    "end": 851.73,
                    "word": " once",
                    "probability": 0.8056640625
                },
                {
                    "start": 851.73,
                    "end": 852.29,
                    "word": ".",
                    "probability": 0.89501953125
                }
            ]
        },
        {
            "id": 114,
            "seek": 87153,
            "start": 852.39,
            "end": 858.33,
            "text": " This prevents specific patterns like delivery worker people from being too revealing.",
            "tokens": [
                50714,
                639,
                22367,
                2685,
                8294,
                411,
                8982,
                11346,
                561,
                490,
                885,
                886,
                23983,
                13,
                51014
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1643350252578425,
            "compression_ratio": 1.4814814814814814,
            "no_speech_prob": 0.0025119781494140625,
            "words": [
                {
                    "start": 852.39,
                    "end": 852.63,
                    "word": " This",
                    "probability": 0.794921875
                },
                {
                    "start": 852.63,
                    "end": 853.05,
                    "word": " prevents",
                    "probability": 0.81201171875
                },
                {
                    "start": 853.05,
                    "end": 853.57,
                    "word": " specific",
                    "probability": 0.80615234375
                },
                {
                    "start": 853.57,
                    "end": 854.15,
                    "word": " patterns",
                    "probability": 0.76904296875
                },
                {
                    "start": 854.15,
                    "end": 854.83,
                    "word": " like",
                    "probability": 0.52490234375
                },
                {
                    "start": 854.83,
                    "end": 855.45,
                    "word": " delivery",
                    "probability": 0.623046875
                },
                {
                    "start": 855.45,
                    "end": 855.83,
                    "word": " worker",
                    "probability": 0.6552734375
                },
                {
                    "start": 855.83,
                    "end": 856.33,
                    "word": " people",
                    "probability": 0.8671875
                },
                {
                    "start": 856.33,
                    "end": 856.89,
                    "word": " from",
                    "probability": 0.7685546875
                },
                {
                    "start": 856.89,
                    "end": 857.13,
                    "word": " being",
                    "probability": 0.87255859375
                },
                {
                    "start": 857.13,
                    "end": 857.45,
                    "word": " too",
                    "probability": 0.8251953125
                },
                {
                    "start": 857.45,
                    "end": 858.05,
                    "word": " revealing",
                    "probability": 0.89013671875
                },
                {
                    "start": 858.05,
                    "end": 858.33,
                    "word": ".",
                    "probability": 0.9658203125
                }
            ]
        },
        {
            "id": 115,
            "seek": 87153,
            "start": 859.49,
            "end": 863.31,
            "text": " Second, we remove metrics that are too noisy.",
            "tokens": [
                51064,
                5736,
                11,
                321,
                4159,
                16367,
                300,
                366,
                886,
                24518,
                13,
                51214
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1643350252578425,
            "compression_ratio": 1.4814814814814814,
            "no_speech_prob": 0.0025119781494140625,
            "words": [
                {
                    "start": 859.49,
                    "end": 860.05,
                    "word": " Second",
                    "probability": 0.75537109375
                },
                {
                    "start": 860.05,
                    "end": 860.33,
                    "word": ",",
                    "probability": 0.958984375
                },
                {
                    "start": 860.33,
                    "end": 860.51,
                    "word": " we",
                    "probability": 0.89111328125
                },
                {
                    "start": 860.51,
                    "end": 860.87,
                    "word": " remove",
                    "probability": 0.63134765625
                },
                {
                    "start": 860.87,
                    "end": 861.33,
                    "word": " metrics",
                    "probability": 0.7861328125
                },
                {
                    "start": 861.33,
                    "end": 861.55,
                    "word": " that",
                    "probability": 0.875
                },
                {
                    "start": 861.55,
                    "end": 861.97,
                    "word": " are",
                    "probability": 0.8720703125
                },
                {
                    "start": 861.97,
                    "end": 862.29,
                    "word": " too",
                    "probability": 0.830078125
                },
                {
                    "start": 862.29,
                    "end": 862.75,
                    "word": " noisy",
                    "probability": 0.74365234375
                },
                {
                    "start": 862.75,
                    "end": 863.31,
                    "word": ".",
                    "probability": 0.916015625
                }
            ]
        },
        {
            "id": 116,
            "seek": 87153,
            "start": 863.45,
            "end": 873.57,
            "text": " By using confidence intervals, we can figure out when a number has a less than 5% chance of being wrong by more than 10%.",
            "tokens": [
                51264,
                3146,
                1228,
                6687,
                26651,
                11,
                321,
                393,
                2573,
                484,
                562,
                257,
                1230,
                575,
                257,
                1570,
                813,
                1025,
                4,
                2931,
                295,
                885,
                2085,
                538,
                544,
                813,
                1266,
                6856,
                51764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1643350252578425,
            "compression_ratio": 1.4814814814814814,
            "no_speech_prob": 0.0025119781494140625,
            "words": [
                {
                    "start": 863.45,
                    "end": 863.65,
                    "word": " By",
                    "probability": 0.87548828125
                },
                {
                    "start": 863.65,
                    "end": 863.99,
                    "word": " using",
                    "probability": 0.80029296875
                },
                {
                    "start": 863.99,
                    "end": 864.49,
                    "word": " confidence",
                    "probability": 0.94189453125
                },
                {
                    "start": 864.49,
                    "end": 864.95,
                    "word": " intervals",
                    "probability": 0.70703125
                },
                {
                    "start": 864.95,
                    "end": 865.09,
                    "word": ",",
                    "probability": 0.67919921875
                },
                {
                    "start": 865.09,
                    "end": 865.21,
                    "word": " we",
                    "probability": 0.90478515625
                },
                {
                    "start": 865.21,
                    "end": 865.41,
                    "word": " can",
                    "probability": 0.9013671875
                },
                {
                    "start": 865.41,
                    "end": 865.69,
                    "word": " figure",
                    "probability": 0.8994140625
                },
                {
                    "start": 865.69,
                    "end": 866.37,
                    "word": " out",
                    "probability": 0.8076171875
                },
                {
                    "start": 866.37,
                    "end": 867.27,
                    "word": " when",
                    "probability": 0.8447265625
                },
                {
                    "start": 867.95,
                    "end": 868.35,
                    "word": " a number",
                    "probability": 0.86181640625
                },
                {
                    "start": 868.35,
                    "end": 869.83,
                    "word": " has",
                    "probability": 0.82568359375
                },
                {
                    "start": 870.03,
                    "end": 870.21,
                    "word": " a less",
                    "probability": 0.84716796875
                },
                {
                    "start": 870.21,
                    "end": 870.33,
                    "word": " than",
                    "probability": 0.861328125
                },
                {
                    "start": 870.33,
                    "end": 870.59,
                    "word": " 5",
                    "probability": 0.8759765625
                },
                {
                    "start": 870.59,
                    "end": 870.79,
                    "word": "%",
                    "probability": 0.9375
                },
                {
                    "start": 870.79,
                    "end": 871.27,
                    "word": " chance",
                    "probability": 0.91845703125
                },
                {
                    "start": 871.27,
                    "end": 871.51,
                    "word": " of",
                    "probability": 0.9033203125
                },
                {
                    "start": 871.51,
                    "end": 871.73,
                    "word": " being",
                    "probability": 0.88232421875
                },
                {
                    "start": 871.73,
                    "end": 871.95,
                    "word": " wrong",
                    "probability": 0.8427734375
                },
                {
                    "start": 871.95,
                    "end": 872.15,
                    "word": " by",
                    "probability": 0.88623046875
                },
                {
                    "start": 872.15,
                    "end": 872.43,
                    "word": " more",
                    "probability": 0.8779296875
                },
                {
                    "start": 872.43,
                    "end": 872.67,
                    "word": " than",
                    "probability": 0.8720703125
                },
                {
                    "start": 872.67,
                    "end": 873.57,
                    "word": " 10%.",
                    "probability": 0.9326171875
                }
            ]
        },
        {
            "id": 117,
            "seek": 90049,
            "start": 873.57,
            "end": 880.63,
            "text": " And finally, if you're curious, we published a technical blog detailing the anonymization process.",
            "tokens": [
                50364,
                400,
                2721,
                11,
                498,
                291,
                434,
                6369,
                11,
                321,
                6572,
                257,
                6191,
                6968,
                42459,
                264,
                37293,
                2144,
                1399,
                13,
                50664
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2663872092962265,
            "compression_ratio": 1.4845814977973568,
            "no_speech_prob": 0.010009765625,
            "words": [
                {
                    "start": 873.57,
                    "end": 874.55,
                    "word": " And",
                    "probability": 0.708984375
                },
                {
                    "start": 874.55,
                    "end": 874.97,
                    "word": " finally",
                    "probability": 0.759765625
                },
                {
                    "start": 874.97,
                    "end": 875.21,
                    "word": ",",
                    "probability": 0.8642578125
                },
                {
                    "start": 875.21,
                    "end": 875.37,
                    "word": " if",
                    "probability": 0.90283203125
                },
                {
                    "start": 875.37,
                    "end": 875.53,
                    "word": " you're",
                    "probability": 0.90478515625
                },
                {
                    "start": 875.53,
                    "end": 875.97,
                    "word": " curious",
                    "probability": 0.7890625
                },
                {
                    "start": 875.97,
                    "end": 876.19,
                    "word": ",",
                    "probability": 0.88134765625
                },
                {
                    "start": 876.19,
                    "end": 876.25,
                    "word": " we",
                    "probability": 0.90087890625
                },
                {
                    "start": 876.25,
                    "end": 876.65,
                    "word": " published",
                    "probability": 0.61767578125
                },
                {
                    "start": 876.85,
                    "end": 877.23,
                    "word": " a technical",
                    "probability": 0.83056640625
                },
                {
                    "start": 877.23,
                    "end": 877.61,
                    "word": " blog",
                    "probability": 0.11505126953125
                },
                {
                    "start": 877.61,
                    "end": 878.25,
                    "word": " detailing",
                    "probability": 0.7177734375
                },
                {
                    "start": 878.25,
                    "end": 878.61,
                    "word": " the",
                    "probability": 0.79638671875
                },
                {
                    "start": 878.61,
                    "end": 879.21,
                    "word": " anonymization",
                    "probability": 0.73779296875
                },
                {
                    "start": 879.21,
                    "end": 879.95,
                    "word": " process",
                    "probability": 0.8720703125
                },
                {
                    "start": 879.95,
                    "end": 880.63,
                    "word": ".",
                    "probability": 0.97900390625
                }
            ]
        },
        {
            "id": 118,
            "seek": 90049,
            "start": 882.57,
            "end": 891.95,
            "text": " Differential privacy has been around for a while, and mobility reports are not the only example out there that's using it.",
            "tokens": [
                50714,
                413,
                12612,
                2549,
                11427,
                575,
                668,
                926,
                337,
                257,
                1339,
                11,
                293,
                16199,
                7122,
                366,
                406,
                264,
                787,
                1365,
                484,
                456,
                300,
                311,
                1228,
                309,
                13,
                51214
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2663872092962265,
            "compression_ratio": 1.4845814977973568,
            "no_speech_prob": 0.010009765625,
            "words": [
                {
                    "start": 882.57,
                    "end": 883.25,
                    "word": " Differential",
                    "probability": 0.7804361979166666
                },
                {
                    "start": 883.25,
                    "end": 883.69,
                    "word": " privacy",
                    "probability": 0.8046875
                },
                {
                    "start": 883.69,
                    "end": 884.39,
                    "word": " has",
                    "probability": 0.7724609375
                },
                {
                    "start": 884.39,
                    "end": 885.23,
                    "word": " been",
                    "probability": 0.81787109375
                },
                {
                    "start": 885.23,
                    "end": 885.53,
                    "word": " around",
                    "probability": 0.86865234375
                },
                {
                    "start": 885.53,
                    "end": 885.77,
                    "word": " for",
                    "probability": 0.900390625
                },
                {
                    "start": 885.87,
                    "end": 886.09,
                    "word": " a while",
                    "probability": 0.91162109375
                },
                {
                    "start": 886.09,
                    "end": 886.21,
                    "word": ",",
                    "probability": 0.341064453125
                },
                {
                    "start": 886.21,
                    "end": 886.25,
                    "word": " and",
                    "probability": 0.88427734375
                },
                {
                    "start": 886.25,
                    "end": 886.59,
                    "word": " mobility",
                    "probability": 0.281982421875
                },
                {
                    "start": 886.59,
                    "end": 887.35,
                    "word": " reports",
                    "probability": 0.75439453125
                },
                {
                    "start": 887.35,
                    "end": 887.63,
                    "word": " are",
                    "probability": 0.89453125
                },
                {
                    "start": 887.63,
                    "end": 887.81,
                    "word": " not",
                    "probability": 0.896484375
                },
                {
                    "start": 887.81,
                    "end": 888.01,
                    "word": " the",
                    "probability": 0.81591796875
                },
                {
                    "start": 888.01,
                    "end": 888.25,
                    "word": " only",
                    "probability": 0.85009765625
                },
                {
                    "start": 888.25,
                    "end": 888.83,
                    "word": " example",
                    "probability": 0.90673828125
                },
                {
                    "start": 888.83,
                    "end": 890.07,
                    "word": " out",
                    "probability": 0.74462890625
                },
                {
                    "start": 890.07,
                    "end": 890.59,
                    "word": " there",
                    "probability": 0.85595703125
                },
                {
                    "start": 890.59,
                    "end": 891.07,
                    "word": " that's",
                    "probability": 0.88427734375
                },
                {
                    "start": 891.07,
                    "end": 891.35,
                    "word": " using",
                    "probability": 0.83203125
                },
                {
                    "start": 891.35,
                    "end": 891.55,
                    "word": " it",
                    "probability": 0.90576171875
                },
                {
                    "start": 891.55,
                    "end": 891.95,
                    "word": ".",
                    "probability": 0.98388671875
                }
            ]
        },
        {
            "id": 119,
            "seek": 90049,
            "start": 891.95,
            "end": 896.05,
            "text": " The US Census is using it for its 2020 census.",
            "tokens": [
                51264,
                440,
                2546,
                34273,
                307,
                1228,
                309,
                337,
                1080,
                4808,
                23725,
                13,
                51414
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2663872092962265,
            "compression_ratio": 1.4845814977973568,
            "no_speech_prob": 0.010009765625,
            "words": [
                {
                    "start": 891.95,
                    "end": 892.07,
                    "word": " The",
                    "probability": 0.8076171875
                },
                {
                    "start": 892.07,
                    "end": 892.27,
                    "word": " US",
                    "probability": 0.7138671875
                },
                {
                    "start": 892.27,
                    "end": 892.59,
                    "word": " Census",
                    "probability": 0.716796875
                },
                {
                    "start": 892.59,
                    "end": 893.89,
                    "word": " is",
                    "probability": 0.8623046875
                },
                {
                    "start": 893.89,
                    "end": 894.19,
                    "word": " using",
                    "probability": 0.8125
                },
                {
                    "start": 894.19,
                    "end": 894.35,
                    "word": " it",
                    "probability": 0.89794921875
                },
                {
                    "start": 894.35,
                    "end": 894.57,
                    "word": " for",
                    "probability": 0.890625
                },
                {
                    "start": 894.57,
                    "end": 894.75,
                    "word": " its",
                    "probability": 0.71923828125
                },
                {
                    "start": 894.75,
                    "end": 895.03,
                    "word": " 2020",
                    "probability": 0.92431640625
                },
                {
                    "start": 895.03,
                    "end": 895.49,
                    "word": " census",
                    "probability": 0.5771484375
                },
                {
                    "start": 895.49,
                    "end": 896.05,
                    "word": ".",
                    "probability": 0.80712890625
                }
            ]
        },
        {
            "id": 120,
            "seek": 90049,
            "start": 896.05,
            "end": 899.45,
            "text": " Facebook recently released an URL data set.",
            "tokens": [
                51464,
                4384,
                3938,
                4736,
                364,
                12905,
                1412,
                992,
                13,
                51614
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2663872092962265,
            "compression_ratio": 1.4845814977973568,
            "no_speech_prob": 0.010009765625,
            "words": [
                {
                    "start": 896.05,
                    "end": 896.31,
                    "word": " Facebook",
                    "probability": 0.7548828125
                },
                {
                    "start": 896.31,
                    "end": 896.81,
                    "word": " recently",
                    "probability": 0.794921875
                },
                {
                    "start": 896.81,
                    "end": 897.41,
                    "word": " released",
                    "probability": 0.87255859375
                },
                {
                    "start": 897.41,
                    "end": 897.63,
                    "word": " an",
                    "probability": 0.67529296875
                },
                {
                    "start": 897.63,
                    "end": 898.03,
                    "word": " URL",
                    "probability": 0.71826171875
                },
                {
                    "start": 898.03,
                    "end": 898.43,
                    "word": " data",
                    "probability": 0.400146484375
                },
                {
                    "start": 898.43,
                    "end": 898.77,
                    "word": " set",
                    "probability": 0.88232421875
                },
                {
                    "start": 898.77,
                    "end": 899.45,
                    "word": ".",
                    "probability": 0.82958984375
                }
            ]
        },
        {
            "id": 121,
            "seek": 90049,
            "start": 899.69,
            "end": 902.53,
            "text": " Apple has been using it.",
            "tokens": [
                51664,
                6373,
                575,
                668,
                1228,
                309,
                13,
                51764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2663872092962265,
            "compression_ratio": 1.4845814977973568,
            "no_speech_prob": 0.010009765625,
            "words": [
                {
                    "start": 899.69,
                    "end": 900.17,
                    "word": " Apple",
                    "probability": 0.8515625
                },
                {
                    "start": 900.17,
                    "end": 900.45,
                    "word": " has",
                    "probability": 0.89453125
                },
                {
                    "start": 900.45,
                    "end": 900.63,
                    "word": " been",
                    "probability": 0.79150390625
                },
                {
                    "start": 900.63,
                    "end": 901.51,
                    "word": " using",
                    "probability": 0.7861328125
                },
                {
                    "start": 901.51,
                    "end": 901.85,
                    "word": " it",
                    "probability": 0.884765625
                },
                {
                    "start": 901.85,
                    "end": 902.53,
                    "word": ".",
                    "probability": 0.65869140625
                }
            ]
        },
        {
            "id": 122,
            "seek": 92915,
            "start": 902.53,
            "end": 904.55,
            "text": " We released it on Chrome a while ago.",
            "tokens": [
                50364,
                492,
                4736,
                309,
                322,
                15327,
                257,
                1339,
                2057,
                13,
                50464
            ],
            "temperature": 0.0,
            "avg_logprob": -0.267795126747202,
            "compression_ratio": 1.6428571428571428,
            "no_speech_prob": 0.12078857421875,
            "words": [
                {
                    "start": 902.53,
                    "end": 902.57,
                    "word": " We",
                    "probability": 0.0036678314208984375
                },
                {
                    "start": 902.57,
                    "end": 902.57,
                    "word": " released",
                    "probability": 0.048583984375
                },
                {
                    "start": 902.57,
                    "end": 902.67,
                    "word": " it",
                    "probability": 0.8671875
                },
                {
                    "start": 902.67,
                    "end": 902.93,
                    "word": " on",
                    "probability": 0.8037109375
                },
                {
                    "start": 902.93,
                    "end": 903.27,
                    "word": " Chrome",
                    "probability": 0.5556640625
                },
                {
                    "start": 903.41,
                    "end": 903.63,
                    "word": " a while",
                    "probability": 0.91357421875
                },
                {
                    "start": 903.63,
                    "end": 903.99,
                    "word": " ago",
                    "probability": 0.80322265625
                },
                {
                    "start": 903.99,
                    "end": 904.55,
                    "word": ".",
                    "probability": 0.258544921875
                }
            ]
        },
        {
            "id": 123,
            "seek": 92915,
            "start": 904.63,
            "end": 909.95,
            "text": " And it's also open source for machine learning purposes on TensorFlow privacy.",
            "tokens": [
                50514,
                400,
                309,
                311,
                611,
                1269,
                4009,
                337,
                3479,
                2539,
                9932,
                322,
                37624,
                11427,
                13,
                50714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.267795126747202,
            "compression_ratio": 1.6428571428571428,
            "no_speech_prob": 0.12078857421875,
            "words": [
                {
                    "start": 904.63,
                    "end": 904.71,
                    "word": " And",
                    "probability": 0.70556640625
                },
                {
                    "start": 904.71,
                    "end": 904.93,
                    "word": " it's",
                    "probability": 0.93359375
                },
                {
                    "start": 904.93,
                    "end": 905.25,
                    "word": " also",
                    "probability": 0.80908203125
                },
                {
                    "start": 905.25,
                    "end": 905.57,
                    "word": " open",
                    "probability": 0.7744140625
                },
                {
                    "start": 905.57,
                    "end": 905.99,
                    "word": " source",
                    "probability": 0.6845703125
                },
                {
                    "start": 905.99,
                    "end": 906.55,
                    "word": " for",
                    "probability": 0.8623046875
                },
                {
                    "start": 906.55,
                    "end": 906.83,
                    "word": " machine",
                    "probability": 0.6904296875
                },
                {
                    "start": 906.83,
                    "end": 907.09,
                    "word": " learning",
                    "probability": 0.8193359375
                },
                {
                    "start": 907.09,
                    "end": 907.55,
                    "word": " purposes",
                    "probability": 0.8935546875
                },
                {
                    "start": 907.55,
                    "end": 908.03,
                    "word": " on",
                    "probability": 0.86474609375
                },
                {
                    "start": 908.03,
                    "end": 908.79,
                    "word": " TensorFlow",
                    "probability": 0.7783203125
                },
                {
                    "start": 908.79,
                    "end": 909.39,
                    "word": " privacy",
                    "probability": 0.5517578125
                },
                {
                    "start": 909.39,
                    "end": 909.95,
                    "word": ".",
                    "probability": 0.9931640625
                }
            ]
        },
        {
            "id": 124,
            "seek": 92915,
            "start": 910.25,
            "end": 915.65,
            "text": " Now, Brian will talk to you a little bit more about building the infrastructure.",
            "tokens": [
                50764,
                823,
                11,
                10765,
                486,
                751,
                281,
                291,
                257,
                707,
                857,
                544,
                466,
                2390,
                264,
                6896,
                13,
                51014
            ],
            "temperature": 0.0,
            "avg_logprob": -0.267795126747202,
            "compression_ratio": 1.6428571428571428,
            "no_speech_prob": 0.12078857421875,
            "words": [
                {
                    "start": 910.25,
                    "end": 910.81,
                    "word": " Now",
                    "probability": 0.89208984375
                },
                {
                    "start": 910.81,
                    "end": 911.23,
                    "word": ",",
                    "probability": 0.794921875
                },
                {
                    "start": 911.23,
                    "end": 911.47,
                    "word": " Brian",
                    "probability": 0.70556640625
                },
                {
                    "start": 911.47,
                    "end": 911.67,
                    "word": " will",
                    "probability": 0.7646484375
                },
                {
                    "start": 911.67,
                    "end": 911.89,
                    "word": " talk",
                    "probability": 0.8076171875
                },
                {
                    "start": 911.89,
                    "end": 912.05,
                    "word": " to",
                    "probability": 0.55908203125
                },
                {
                    "start": 912.05,
                    "end": 912.27,
                    "word": " you",
                    "probability": 0.63232421875
                },
                {
                    "start": 912.39,
                    "end": 912.55,
                    "word": " a little",
                    "probability": 0.80712890625
                },
                {
                    "start": 912.55,
                    "end": 912.87,
                    "word": " bit",
                    "probability": 0.88330078125
                },
                {
                    "start": 912.87,
                    "end": 913.55,
                    "word": " more",
                    "probability": 0.85888671875
                },
                {
                    "start": 913.55,
                    "end": 913.83,
                    "word": " about",
                    "probability": 0.7919921875
                },
                {
                    "start": 913.83,
                    "end": 914.25,
                    "word": " building",
                    "probability": 0.810546875
                },
                {
                    "start": 914.25,
                    "end": 914.51,
                    "word": " the",
                    "probability": 0.810546875
                },
                {
                    "start": 914.51,
                    "end": 915.09,
                    "word": " infrastructure",
                    "probability": 0.86669921875
                },
                {
                    "start": 915.09,
                    "end": 915.65,
                    "word": ".",
                    "probability": 0.99365234375
                }
            ]
        },
        {
            "id": 125,
            "seek": 92915,
            "start": 915.83,
            "end": 922.29,
            "text": " So getting to the point where we could build a new product from the ground up with differential privacy was a long journey.",
            "tokens": [
                51064,
                407,
                1242,
                281,
                264,
                935,
                689,
                321,
                727,
                1322,
                257,
                777,
                1674,
                490,
                264,
                2727,
                493,
                365,
                15756,
                11427,
                390,
                257,
                938,
                4671,
                13,
                51314
            ],
            "temperature": 0.0,
            "avg_logprob": -0.267795126747202,
            "compression_ratio": 1.6428571428571428,
            "no_speech_prob": 0.12078857421875,
            "words": [
                {
                    "start": 915.83,
                    "end": 916.39,
                    "word": " So",
                    "probability": 0.7451171875
                },
                {
                    "start": 916.39,
                    "end": 916.61,
                    "word": " getting",
                    "probability": 0.75390625
                },
                {
                    "start": 916.61,
                    "end": 916.75,
                    "word": " to",
                    "probability": 0.90087890625
                },
                {
                    "start": 916.75,
                    "end": 916.87,
                    "word": " the",
                    "probability": 0.81640625
                },
                {
                    "start": 916.87,
                    "end": 917.03,
                    "word": " point",
                    "probability": 0.85498046875
                },
                {
                    "start": 917.03,
                    "end": 917.19,
                    "word": " where",
                    "probability": 0.90673828125
                },
                {
                    "start": 917.19,
                    "end": 917.33,
                    "word": " we",
                    "probability": 0.904296875
                },
                {
                    "start": 917.33,
                    "end": 917.47,
                    "word": " could",
                    "probability": 0.75439453125
                },
                {
                    "start": 917.47,
                    "end": 917.83,
                    "word": " build",
                    "probability": 0.86669921875
                },
                {
                    "start": 918.25,
                    "end": 918.43,
                    "word": " a new",
                    "probability": 0.82763671875
                },
                {
                    "start": 918.43,
                    "end": 918.87,
                    "word": " product",
                    "probability": 0.7958984375
                },
                {
                    "start": 918.87,
                    "end": 919.13,
                    "word": " from",
                    "probability": 0.806640625
                },
                {
                    "start": 919.13,
                    "end": 919.29,
                    "word": " the",
                    "probability": 0.81298828125
                },
                {
                    "start": 919.29,
                    "end": 919.57,
                    "word": " ground",
                    "probability": 0.87646484375
                },
                {
                    "start": 919.57,
                    "end": 919.83,
                    "word": " up",
                    "probability": 0.90234375
                },
                {
                    "start": 919.83,
                    "end": 920.01,
                    "word": " with",
                    "probability": 0.78955078125
                },
                {
                    "start": 920.01,
                    "end": 920.33,
                    "word": " differential",
                    "probability": 0.63232421875
                },
                {
                    "start": 920.33,
                    "end": 920.77,
                    "word": " privacy",
                    "probability": 0.84423828125
                },
                {
                    "start": 920.77,
                    "end": 921.05,
                    "word": " was",
                    "probability": 0.8857421875
                },
                {
                    "start": 921.23,
                    "end": 921.65,
                    "word": " a long",
                    "probability": 0.89111328125
                },
                {
                    "start": 921.65,
                    "end": 922.03,
                    "word": " journey",
                    "probability": 0.857421875
                },
                {
                    "start": 922.03,
                    "end": 922.29,
                    "word": ".",
                    "probability": 0.759765625
                }
            ]
        },
        {
            "id": 126,
            "seek": 92915,
            "start": 922.29,
            "end": 928.15,
            "text": " But I want to take you through three really critically important pieces of that journey, which is building strong foundations,",
            "tokens": [
                51364,
                583,
                286,
                528,
                281,
                747,
                291,
                807,
                1045,
                534,
                22797,
                1021,
                3755,
                295,
                300,
                4671,
                11,
                597,
                307,
                2390,
                2068,
                22467,
                11,
                51614
            ],
            "temperature": 0.0,
            "avg_logprob": -0.267795126747202,
            "compression_ratio": 1.6428571428571428,
            "no_speech_prob": 0.12078857421875,
            "words": [
                {
                    "start": 922.29,
                    "end": 922.53,
                    "word": " But",
                    "probability": 0.87841796875
                },
                {
                    "start": 922.53,
                    "end": 922.61,
                    "word": " I",
                    "probability": 0.9951171875
                },
                {
                    "start": 922.61,
                    "end": 922.71,
                    "word": " want",
                    "probability": 0.78125
                },
                {
                    "start": 922.71,
                    "end": 922.79,
                    "word": " to",
                    "probability": 0.904296875
                },
                {
                    "start": 922.79,
                    "end": 922.97,
                    "word": " take",
                    "probability": 0.7939453125
                },
                {
                    "start": 922.97,
                    "end": 923.11,
                    "word": " you",
                    "probability": 0.89599609375
                },
                {
                    "start": 923.11,
                    "end": 923.47,
                    "word": " through",
                    "probability": 0.7802734375
                },
                {
                    "start": 923.47,
                    "end": 924.03,
                    "word": " three",
                    "probability": 0.8349609375
                },
                {
                    "start": 924.03,
                    "end": 924.33,
                    "word": " really",
                    "probability": 0.7724609375
                },
                {
                    "start": 924.33,
                    "end": 924.65,
                    "word": " critically",
                    "probability": 0.206787109375
                },
                {
                    "start": 924.65,
                    "end": 925.05,
                    "word": " important",
                    "probability": 0.7509765625
                },
                {
                    "start": 925.05,
                    "end": 925.49,
                    "word": " pieces",
                    "probability": 0.87841796875
                },
                {
                    "start": 925.49,
                    "end": 925.67,
                    "word": " of",
                    "probability": 0.9052734375
                },
                {
                    "start": 925.67,
                    "end": 925.85,
                    "word": " that",
                    "probability": 0.880859375
                },
                {
                    "start": 925.85,
                    "end": 926.09,
                    "word": " journey",
                    "probability": 0.85107421875
                },
                {
                    "start": 926.09,
                    "end": 926.21,
                    "word": ",",
                    "probability": 0.8720703125
                },
                {
                    "start": 926.21,
                    "end": 926.37,
                    "word": " which",
                    "probability": 0.92138671875
                },
                {
                    "start": 926.37,
                    "end": 926.51,
                    "word": " is",
                    "probability": 0.89990234375
                },
                {
                    "start": 926.51,
                    "end": 926.75,
                    "word": " building",
                    "probability": 0.82568359375
                },
                {
                    "start": 926.75,
                    "end": 927.23,
                    "word": " strong",
                    "probability": 0.82763671875
                },
                {
                    "start": 927.23,
                    "end": 927.77,
                    "word": " foundations",
                    "probability": 0.9052734375
                },
                {
                    "start": 927.77,
                    "end": 928.15,
                    "word": ",",
                    "probability": 0.97021484375
                }
            ]
        },
        {
            "id": 127,
            "seek": 92915,
            "start": 928.15,
            "end": 931.19,
            "text": " building frameworks that are familiar to potential users,",
            "tokens": [
                51664,
                2390,
                29834,
                300,
                366,
                4963,
                281,
                3995,
                5022,
                11,
                51764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.267795126747202,
            "compression_ratio": 1.6428571428571428,
            "no_speech_prob": 0.12078857421875,
            "words": [
                {
                    "start": 928.15,
                    "end": 928.43,
                    "word": " building",
                    "probability": 0.80859375
                },
                {
                    "start": 928.43,
                    "end": 928.89,
                    "word": " frameworks",
                    "probability": 0.6650390625
                },
                {
                    "start": 928.89,
                    "end": 929.13,
                    "word": " that",
                    "probability": 0.876953125
                },
                {
                    "start": 929.13,
                    "end": 929.23,
                    "word": " are",
                    "probability": 0.8896484375
                },
                {
                    "start": 929.23,
                    "end": 929.57,
                    "word": " familiar",
                    "probability": 0.76416015625
                },
                {
                    "start": 929.57,
                    "end": 929.87,
                    "word": " to",
                    "probability": 0.9072265625
                },
                {
                    "start": 929.87,
                    "end": 930.23,
                    "word": " potential",
                    "probability": 0.853515625
                },
                {
                    "start": 930.23,
                    "end": 930.69,
                    "word": " users",
                    "probability": 0.8388671875
                },
                {
                    "start": 930.69,
                    "end": 931.19,
                    "word": ",",
                    "probability": 0.900390625
                }
            ]
        },
        {
            "id": 128,
            "seek": 95545,
            "start": 931.19,
            "end": 935.19,
            "text": " and helping users to assess the privacy utility frontier.",
            "tokens": [
                50364,
                293,
                4315,
                5022,
                281,
                5877,
                264,
                11427,
                14877,
                35853,
                13,
                50514
            ],
            "temperature": 0.0,
            "avg_logprob": -0.21776108273976966,
            "compression_ratio": 1.6762295081967213,
            "no_speech_prob": 0.0017824172973632812,
            "words": [
                {
                    "start": 931.19,
                    "end": 931.59,
                    "word": " and",
                    "probability": 0.7802734375
                },
                {
                    "start": 931.59,
                    "end": 931.97,
                    "word": " helping",
                    "probability": 0.72216796875
                },
                {
                    "start": 931.97,
                    "end": 932.37,
                    "word": " users",
                    "probability": 0.82861328125
                },
                {
                    "start": 932.37,
                    "end": 932.59,
                    "word": " to",
                    "probability": 0.892578125
                },
                {
                    "start": 932.59,
                    "end": 932.89,
                    "word": " assess",
                    "probability": 0.88671875
                },
                {
                    "start": 932.89,
                    "end": 933.11,
                    "word": " the",
                    "probability": 0.82080078125
                },
                {
                    "start": 933.11,
                    "end": 933.45,
                    "word": " privacy",
                    "probability": 0.83056640625
                },
                {
                    "start": 933.45,
                    "end": 934.07,
                    "word": " utility",
                    "probability": 0.76123046875
                },
                {
                    "start": 934.07,
                    "end": 934.51,
                    "word": " frontier",
                    "probability": 0.87890625
                },
                {
                    "start": 934.51,
                    "end": 935.19,
                    "word": ".",
                    "probability": 0.98046875
                }
            ]
        },
        {
            "id": 129,
            "seek": 95545,
            "start": 935.49,
            "end": 941.27,
            "text": " So the infrastructure that we build starts with the differentially private foundational libraries.",
            "tokens": [
                50564,
                407,
                264,
                6896,
                300,
                321,
                1322,
                3719,
                365,
                264,
                819,
                2270,
                4551,
                32195,
                15148,
                13,
                50814
            ],
            "temperature": 0.0,
            "avg_logprob": -0.21776108273976966,
            "compression_ratio": 1.6762295081967213,
            "no_speech_prob": 0.0017824172973632812,
            "words": [
                {
                    "start": 935.49,
                    "end": 936.09,
                    "word": " So",
                    "probability": 0.814453125
                },
                {
                    "start": 936.09,
                    "end": 936.61,
                    "word": " the",
                    "probability": 0.56640625
                },
                {
                    "start": 936.61,
                    "end": 937.15,
                    "word": " infrastructure",
                    "probability": 0.88330078125
                },
                {
                    "start": 937.15,
                    "end": 937.43,
                    "word": " that",
                    "probability": 0.88427734375
                },
                {
                    "start": 937.43,
                    "end": 937.55,
                    "word": " we",
                    "probability": 0.9072265625
                },
                {
                    "start": 937.55,
                    "end": 937.81,
                    "word": " build",
                    "probability": 0.6494140625
                },
                {
                    "start": 937.81,
                    "end": 938.15,
                    "word": " starts",
                    "probability": 0.76611328125
                },
                {
                    "start": 938.15,
                    "end": 938.41,
                    "word": " with",
                    "probability": 0.81689453125
                },
                {
                    "start": 938.41,
                    "end": 938.63,
                    "word": " the",
                    "probability": 0.6572265625
                },
                {
                    "start": 938.63,
                    "end": 939.05,
                    "word": " differentially",
                    "probability": 0.762451171875
                },
                {
                    "start": 939.05,
                    "end": 939.47,
                    "word": " private",
                    "probability": 0.7890625
                },
                {
                    "start": 939.47,
                    "end": 940.01,
                    "word": " foundational",
                    "probability": 0.8662109375
                },
                {
                    "start": 940.01,
                    "end": 940.59,
                    "word": " libraries",
                    "probability": 0.900390625
                },
                {
                    "start": 940.59,
                    "end": 941.27,
                    "word": ".",
                    "probability": 0.9501953125
                }
            ]
        },
        {
            "id": 130,
            "seek": 95545,
            "start": 941.29,
            "end": 950.29,
            "text": " We started these about three years ago, and our vision was to power differentially private use cases that could primarily accelerate developer velocity.",
            "tokens": [
                50864,
                492,
                1409,
                613,
                466,
                1045,
                924,
                2057,
                11,
                293,
                527,
                5201,
                390,
                281,
                1347,
                819,
                2270,
                4551,
                764,
                3331,
                300,
                727,
                10029,
                21341,
                10754,
                9269,
                13,
                51264
            ],
            "temperature": 0.0,
            "avg_logprob": -0.21776108273976966,
            "compression_ratio": 1.6762295081967213,
            "no_speech_prob": 0.0017824172973632812,
            "words": [
                {
                    "start": 941.29,
                    "end": 941.55,
                    "word": " We",
                    "probability": 0.8916015625
                },
                {
                    "start": 941.55,
                    "end": 941.79,
                    "word": " started",
                    "probability": 0.7958984375
                },
                {
                    "start": 941.79,
                    "end": 942.11,
                    "word": " these",
                    "probability": 0.7197265625
                },
                {
                    "start": 942.11,
                    "end": 942.39,
                    "word": " about",
                    "probability": 0.81591796875
                },
                {
                    "start": 942.39,
                    "end": 942.67,
                    "word": " three",
                    "probability": 0.82373046875
                },
                {
                    "start": 942.67,
                    "end": 942.95,
                    "word": " years",
                    "probability": 0.88037109375
                },
                {
                    "start": 942.95,
                    "end": 943.31,
                    "word": " ago",
                    "probability": 0.80419921875
                },
                {
                    "start": 943.31,
                    "end": 943.49,
                    "word": ",",
                    "probability": 0.47998046875
                },
                {
                    "start": 943.49,
                    "end": 943.83,
                    "word": " and",
                    "probability": 0.89453125
                },
                {
                    "start": 943.83,
                    "end": 944.05,
                    "word": " our",
                    "probability": 0.80810546875
                },
                {
                    "start": 944.05,
                    "end": 944.39,
                    "word": " vision",
                    "probability": 0.80029296875
                },
                {
                    "start": 944.39,
                    "end": 944.73,
                    "word": " was",
                    "probability": 0.8984375
                },
                {
                    "start": 944.73,
                    "end": 944.91,
                    "word": " to",
                    "probability": 0.8896484375
                },
                {
                    "start": 944.91,
                    "end": 945.23,
                    "word": " power",
                    "probability": 0.80029296875
                },
                {
                    "start": 945.23,
                    "end": 945.77,
                    "word": " differentially",
                    "probability": 0.865478515625
                },
                {
                    "start": 945.77,
                    "end": 946.09,
                    "word": " private",
                    "probability": 0.80859375
                },
                {
                    "start": 946.09,
                    "end": 946.43,
                    "word": " use",
                    "probability": 0.80419921875
                },
                {
                    "start": 946.43,
                    "end": 946.89,
                    "word": " cases",
                    "probability": 0.82470703125
                },
                {
                    "start": 946.89,
                    "end": 947.45,
                    "word": " that",
                    "probability": 0.83935546875
                },
                {
                    "start": 947.45,
                    "end": 947.73,
                    "word": " could",
                    "probability": 0.80322265625
                },
                {
                    "start": 947.73,
                    "end": 948.19,
                    "word": " primarily",
                    "probability": 0.833984375
                },
                {
                    "start": 948.19,
                    "end": 948.79,
                    "word": " accelerate",
                    "probability": 0.63916015625
                },
                {
                    "start": 948.79,
                    "end": 949.23,
                    "word": " developer",
                    "probability": 0.63818359375
                },
                {
                    "start": 949.23,
                    "end": 949.75,
                    "word": " velocity",
                    "probability": 0.6767578125
                },
                {
                    "start": 949.75,
                    "end": 950.29,
                    "word": ".",
                    "probability": 0.9931640625
                }
            ]
        },
        {
            "id": 131,
            "seek": 95545,
            "start": 950.29,
            "end": 957.49,
            "text": " A lot of times users are trying to do experiments or prototypes before actually building a project,",
            "tokens": [
                51314,
                316,
                688,
                295,
                1413,
                5022,
                366,
                1382,
                281,
                360,
                12050,
                420,
                42197,
                949,
                767,
                2390,
                257,
                1716,
                11,
                51664
            ],
            "temperature": 0.0,
            "avg_logprob": -0.21776108273976966,
            "compression_ratio": 1.6762295081967213,
            "no_speech_prob": 0.0017824172973632812,
            "words": [
                {
                    "start": 950.29,
                    "end": 950.75,
                    "word": " A",
                    "probability": 0.9951171875
                },
                {
                    "start": 950.75,
                    "end": 950.89,
                    "word": " lot",
                    "probability": 0.88818359375
                },
                {
                    "start": 950.89,
                    "end": 950.99,
                    "word": " of",
                    "probability": 0.904296875
                },
                {
                    "start": 950.99,
                    "end": 951.31,
                    "word": " times",
                    "probability": 0.8173828125
                },
                {
                    "start": 951.31,
                    "end": 951.59,
                    "word": " users",
                    "probability": 0.6455078125
                },
                {
                    "start": 951.59,
                    "end": 952.15,
                    "word": " are",
                    "probability": 0.89892578125
                },
                {
                    "start": 952.15,
                    "end": 952.63,
                    "word": " trying",
                    "probability": 0.81103515625
                },
                {
                    "start": 952.63,
                    "end": 952.87,
                    "word": " to",
                    "probability": 0.90966796875
                },
                {
                    "start": 952.87,
                    "end": 953.31,
                    "word": " do",
                    "probability": 0.9013671875
                },
                {
                    "start": 953.31,
                    "end": 954.09,
                    "word": " experiments",
                    "probability": 0.88818359375
                },
                {
                    "start": 954.09,
                    "end": 954.45,
                    "word": " or",
                    "probability": 0.88720703125
                },
                {
                    "start": 954.45,
                    "end": 955.05,
                    "word": " prototypes",
                    "probability": 0.73681640625
                },
                {
                    "start": 955.05,
                    "end": 956.35,
                    "word": " before",
                    "probability": 0.751953125
                },
                {
                    "start": 956.35,
                    "end": 956.65,
                    "word": " actually",
                    "probability": 0.77197265625
                },
                {
                    "start": 956.65,
                    "end": 956.91,
                    "word": " building",
                    "probability": 0.77880859375
                },
                {
                    "start": 957.07,
                    "end": 957.33,
                    "word": " a project",
                    "probability": 0.83349609375
                },
                {
                    "start": 957.33,
                    "end": 957.49,
                    "word": ",",
                    "probability": 0.464111328125
                }
            ]
        },
        {
            "id": 132,
            "seek": 98545,
            "start": 957.49,
            "end": 961.17,
            "text": " and this ideation stage is perfect for differential privacy,",
            "tokens": [
                50364,
                293,
                341,
                1153,
                399,
                3233,
                307,
                2176,
                337,
                15756,
                11427,
                11,
                50514
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1918569662823127,
            "compression_ratio": 1.658703071672355,
            "no_speech_prob": 0.0078125,
            "words": [
                {
                    "start": 957.49,
                    "end": 957.63,
                    "word": " and",
                    "probability": 0.1676025390625
                },
                {
                    "start": 957.63,
                    "end": 957.81,
                    "word": " this",
                    "probability": 0.8740234375
                },
                {
                    "start": 957.81,
                    "end": 958.53,
                    "word": " ideation",
                    "probability": 0.87158203125
                },
                {
                    "start": 958.53,
                    "end": 959.07,
                    "word": " stage",
                    "probability": 0.73974609375
                },
                {
                    "start": 959.07,
                    "end": 959.27,
                    "word": " is",
                    "probability": 0.90478515625
                },
                {
                    "start": 959.27,
                    "end": 959.75,
                    "word": " perfect",
                    "probability": 0.8466796875
                },
                {
                    "start": 959.75,
                    "end": 960.03,
                    "word": " for",
                    "probability": 0.8974609375
                },
                {
                    "start": 960.03,
                    "end": 960.39,
                    "word": " differential",
                    "probability": 0.697265625
                },
                {
                    "start": 960.39,
                    "end": 960.85,
                    "word": " privacy",
                    "probability": 0.84765625
                },
                {
                    "start": 960.85,
                    "end": 961.17,
                    "word": ",",
                    "probability": 0.308837890625
                }
            ]
        },
        {
            "id": 133,
            "seek": 98545,
            "start": 961.17,
            "end": 965.87,
            "text": " because it can provide a very good way of getting trends or overall statistics,",
            "tokens": [
                50564,
                570,
                309,
                393,
                2893,
                257,
                588,
                665,
                636,
                295,
                1242,
                13892,
                420,
                4787,
                12523,
                11,
                50714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1918569662823127,
            "compression_ratio": 1.658703071672355,
            "no_speech_prob": 0.0078125,
            "words": [
                {
                    "start": 961.17,
                    "end": 961.69,
                    "word": " because",
                    "probability": 0.822265625
                },
                {
                    "start": 961.69,
                    "end": 961.87,
                    "word": " it",
                    "probability": 0.90380859375
                },
                {
                    "start": 961.87,
                    "end": 961.99,
                    "word": " can",
                    "probability": 0.8935546875
                },
                {
                    "start": 961.99,
                    "end": 962.27,
                    "word": " provide",
                    "probability": 0.89990234375
                },
                {
                    "start": 962.43,
                    "end": 962.61,
                    "word": " a very",
                    "probability": 0.7412109375
                },
                {
                    "start": 962.61,
                    "end": 962.79,
                    "word": " good",
                    "probability": 0.87890625
                },
                {
                    "start": 962.79,
                    "end": 963.11,
                    "word": " way",
                    "probability": 0.8984375
                },
                {
                    "start": 963.11,
                    "end": 963.39,
                    "word": " of",
                    "probability": 0.90576171875
                },
                {
                    "start": 963.39,
                    "end": 963.63,
                    "word": " getting",
                    "probability": 0.90771484375
                },
                {
                    "start": 963.63,
                    "end": 964.17,
                    "word": " trends",
                    "probability": 0.7255859375
                },
                {
                    "start": 964.17,
                    "end": 964.35,
                    "word": " or",
                    "probability": 0.841796875
                },
                {
                    "start": 964.35,
                    "end": 964.65,
                    "word": " overall",
                    "probability": 0.8623046875
                },
                {
                    "start": 964.65,
                    "end": 965.33,
                    "word": " statistics",
                    "probability": 0.90185546875
                },
                {
                    "start": 965.33,
                    "end": 965.87,
                    "word": ",",
                    "probability": 0.515625
                }
            ]
        },
        {
            "id": 134,
            "seek": 98545,
            "start": 965.87,
            "end": 972.51,
            "text": " and allows people to assess whether it's even worth continuing development on a particular project.",
            "tokens": [
                50764,
                293,
                4045,
                561,
                281,
                5877,
                1968,
                309,
                311,
                754,
                3163,
                9289,
                3250,
                322,
                257,
                1729,
                1716,
                13,
                51064
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1918569662823127,
            "compression_ratio": 1.658703071672355,
            "no_speech_prob": 0.0078125,
            "words": [
                {
                    "start": 965.87,
                    "end": 966.23,
                    "word": " and",
                    "probability": 0.85888671875
                },
                {
                    "start": 966.23,
                    "end": 966.53,
                    "word": " allows",
                    "probability": 0.796875
                },
                {
                    "start": 966.53,
                    "end": 966.77,
                    "word": " people",
                    "probability": 0.93017578125
                },
                {
                    "start": 966.77,
                    "end": 966.95,
                    "word": " to",
                    "probability": 0.90673828125
                },
                {
                    "start": 966.95,
                    "end": 967.31,
                    "word": " assess",
                    "probability": 0.8896484375
                },
                {
                    "start": 967.31,
                    "end": 967.55,
                    "word": " whether",
                    "probability": 0.833984375
                },
                {
                    "start": 967.55,
                    "end": 967.73,
                    "word": " it's",
                    "probability": 0.94921875
                },
                {
                    "start": 967.73,
                    "end": 967.91,
                    "word": " even",
                    "probability": 0.8134765625
                },
                {
                    "start": 967.91,
                    "end": 968.33,
                    "word": " worth",
                    "probability": 0.87255859375
                },
                {
                    "start": 968.33,
                    "end": 970.15,
                    "word": " continuing",
                    "probability": 0.3349609375
                },
                {
                    "start": 970.15,
                    "end": 970.69,
                    "word": " development",
                    "probability": 0.78125
                },
                {
                    "start": 970.69,
                    "end": 970.89,
                    "word": " on",
                    "probability": 0.90625
                },
                {
                    "start": 971.01,
                    "end": 971.33,
                    "word": " a particular",
                    "probability": 0.80712890625
                },
                {
                    "start": 971.33,
                    "end": 971.95,
                    "word": " project",
                    "probability": 0.83349609375
                },
                {
                    "start": 971.95,
                    "end": 972.51,
                    "word": ".",
                    "probability": 0.9951171875
                }
            ]
        },
        {
            "id": 135,
            "seek": 98545,
            "start": 973.47,
            "end": 980.97,
            "text": " So the foundational library that we use here at Google is actually the same as the one that we open sourced and published,",
            "tokens": [
                51114,
                407,
                264,
                32195,
                6405,
                300,
                321,
                764,
                510,
                412,
                3329,
                307,
                767,
                264,
                912,
                382,
                264,
                472,
                300,
                321,
                1269,
                11006,
                1232,
                293,
                6572,
                11,
                51464
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1918569662823127,
            "compression_ratio": 1.658703071672355,
            "no_speech_prob": 0.0078125,
            "words": [
                {
                    "start": 973.47,
                    "end": 973.95,
                    "word": " So",
                    "probability": 0.81640625
                },
                {
                    "start": 973.95,
                    "end": 974.41,
                    "word": " the",
                    "probability": 0.60400390625
                },
                {
                    "start": 974.41,
                    "end": 974.87,
                    "word": " foundational",
                    "probability": 0.603515625
                },
                {
                    "start": 974.87,
                    "end": 975.25,
                    "word": " library",
                    "probability": 0.82275390625
                },
                {
                    "start": 975.25,
                    "end": 975.41,
                    "word": " that",
                    "probability": 0.8798828125
                },
                {
                    "start": 975.41,
                    "end": 975.51,
                    "word": " we",
                    "probability": 0.904296875
                },
                {
                    "start": 975.51,
                    "end": 975.71,
                    "word": " use",
                    "probability": 0.82568359375
                },
                {
                    "start": 975.71,
                    "end": 975.87,
                    "word": " here",
                    "probability": 0.82275390625
                },
                {
                    "start": 975.87,
                    "end": 976.03,
                    "word": " at",
                    "probability": 0.87841796875
                },
                {
                    "start": 976.03,
                    "end": 976.53,
                    "word": " Google",
                    "probability": 0.875
                },
                {
                    "start": 976.53,
                    "end": 976.85,
                    "word": " is",
                    "probability": 0.876953125
                },
                {
                    "start": 976.85,
                    "end": 977.13,
                    "word": " actually",
                    "probability": 0.78271484375
                },
                {
                    "start": 977.13,
                    "end": 977.41,
                    "word": " the",
                    "probability": 0.8134765625
                },
                {
                    "start": 977.41,
                    "end": 977.85,
                    "word": " same",
                    "probability": 0.82421875
                },
                {
                    "start": 977.85,
                    "end": 978.31,
                    "word": " as",
                    "probability": 0.9033203125
                },
                {
                    "start": 978.31,
                    "end": 978.49,
                    "word": " the",
                    "probability": 0.81591796875
                },
                {
                    "start": 978.49,
                    "end": 978.63,
                    "word": " one",
                    "probability": 0.82080078125
                },
                {
                    "start": 978.63,
                    "end": 978.77,
                    "word": " that",
                    "probability": 0.87939453125
                },
                {
                    "start": 978.77,
                    "end": 978.99,
                    "word": " we",
                    "probability": 0.9013671875
                },
                {
                    "start": 978.99,
                    "end": 979.33,
                    "word": " open",
                    "probability": 0.7666015625
                },
                {
                    "start": 979.33,
                    "end": 979.79,
                    "word": " sourced",
                    "probability": 0.802978515625
                },
                {
                    "start": 979.79,
                    "end": 980.01,
                    "word": " and",
                    "probability": 0.89453125
                },
                {
                    "start": 980.01,
                    "end": 980.47,
                    "word": " published",
                    "probability": 0.7080078125
                },
                {
                    "start": 980.47,
                    "end": 980.97,
                    "word": ",",
                    "probability": 0.51708984375
                }
            ]
        },
        {
            "id": 136,
            "seek": 98545,
            "start": 980.97,
            "end": 983.15,
            "text": " so that others could reuse our work.",
            "tokens": [
                51514,
                370,
                300,
                2357,
                727,
                26225,
                527,
                589,
                13,
                51614
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1918569662823127,
            "compression_ratio": 1.658703071672355,
            "no_speech_prob": 0.0078125,
            "words": [
                {
                    "start": 980.97,
                    "end": 981.49,
                    "word": " so",
                    "probability": 0.9072265625
                },
                {
                    "start": 981.49,
                    "end": 981.61,
                    "word": " that",
                    "probability": 0.88134765625
                },
                {
                    "start": 981.61,
                    "end": 981.83,
                    "word": " others",
                    "probability": 0.767578125
                },
                {
                    "start": 981.83,
                    "end": 981.99,
                    "word": " could",
                    "probability": 0.78076171875
                },
                {
                    "start": 981.99,
                    "end": 982.23,
                    "word": " reuse",
                    "probability": 0.77099609375
                },
                {
                    "start": 982.23,
                    "end": 982.45,
                    "word": " our",
                    "probability": 0.8212890625
                },
                {
                    "start": 982.45,
                    "end": 982.75,
                    "word": " work",
                    "probability": 0.8134765625
                },
                {
                    "start": 982.75,
                    "end": 983.15,
                    "word": ".",
                    "probability": 0.99072265625
                }
            ]
        },
        {
            "id": 137,
            "seek": 98545,
            "start": 983.15,
            "end": 987.31,
            "text": " For us, it's always been important to provide transparency around our implementation.",
            "tokens": [
                51664,
                1171,
                505,
                11,
                309,
                311,
                1009,
                668,
                1021,
                281,
                2893,
                17131,
                926,
                527,
                11420,
                13,
                51814
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1918569662823127,
            "compression_ratio": 1.658703071672355,
            "no_speech_prob": 0.0078125,
            "words": [
                {
                    "start": 983.15,
                    "end": 983.51,
                    "word": " For",
                    "probability": 0.8984375
                },
                {
                    "start": 983.51,
                    "end": 983.69,
                    "word": " us",
                    "probability": 0.9072265625
                },
                {
                    "start": 983.69,
                    "end": 983.73,
                    "word": ",",
                    "probability": 0.89892578125
                },
                {
                    "start": 983.73,
                    "end": 983.99,
                    "word": " it's",
                    "probability": 0.949462890625
                },
                {
                    "start": 983.99,
                    "end": 984.27,
                    "word": " always",
                    "probability": 0.779296875
                },
                {
                    "start": 984.27,
                    "end": 984.47,
                    "word": " been",
                    "probability": 0.80810546875
                },
                {
                    "start": 984.47,
                    "end": 984.75,
                    "word": " important",
                    "probability": 0.7705078125
                },
                {
                    "start": 984.75,
                    "end": 984.97,
                    "word": " to",
                    "probability": 0.9072265625
                },
                {
                    "start": 984.97,
                    "end": 985.23,
                    "word": " provide",
                    "probability": 0.89404296875
                },
                {
                    "start": 985.23,
                    "end": 985.91,
                    "word": " transparency",
                    "probability": 0.7080078125
                },
                {
                    "start": 985.91,
                    "end": 986.27,
                    "word": " around",
                    "probability": 0.880859375
                },
                {
                    "start": 986.27,
                    "end": 986.45,
                    "word": " our",
                    "probability": 0.80322265625
                },
                {
                    "start": 986.45,
                    "end": 986.83,
                    "word": " implementation",
                    "probability": 0.70068359375
                },
                {
                    "start": 986.83,
                    "end": 987.31,
                    "word": ".",
                    "probability": 0.8974609375
                }
            ]
        },
        {
            "id": 138,
            "seek": 101253,
            "start": 988.07,
            "end": 991.71,
            "text": " Hopefully folks have had a chance to poke around the library.",
            "tokens": [
                50364,
                10429,
                4024,
                362,
                632,
                257,
                2931,
                281,
                19712,
                926,
                264,
                6405,
                13,
                50564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19977678908478647,
            "compression_ratio": 1.5560344827586208,
            "no_speech_prob": 0.0149383544921875,
            "words": [
                {
                    "start": 988.07,
                    "end": 988.71,
                    "word": " Hopefully",
                    "probability": 0.404541015625
                },
                {
                    "start": 988.71,
                    "end": 989.17,
                    "word": " folks",
                    "probability": 0.61767578125
                },
                {
                    "start": 989.17,
                    "end": 989.33,
                    "word": " have",
                    "probability": 0.876953125
                },
                {
                    "start": 989.33,
                    "end": 989.53,
                    "word": " had",
                    "probability": 0.80517578125
                },
                {
                    "start": 989.67,
                    "end": 990.01,
                    "word": " a chance",
                    "probability": 0.9072265625
                },
                {
                    "start": 990.01,
                    "end": 990.33,
                    "word": " to",
                    "probability": 0.89892578125
                },
                {
                    "start": 990.33,
                    "end": 990.59,
                    "word": " poke",
                    "probability": 0.83544921875
                },
                {
                    "start": 990.59,
                    "end": 990.81,
                    "word": " around",
                    "probability": 0.87890625
                },
                {
                    "start": 990.81,
                    "end": 990.99,
                    "word": " the",
                    "probability": 0.79931640625
                },
                {
                    "start": 990.99,
                    "end": 991.39,
                    "word": " library",
                    "probability": 0.8173828125
                },
                {
                    "start": 991.39,
                    "end": 991.71,
                    "word": ".",
                    "probability": 0.7880859375
                }
            ]
        },
        {
            "id": 139,
            "seek": 101253,
            "start": 992.69,
            "end": 999.71,
            "text": " It was launched in C++ several months ago, and a couple of weeks ago we also launched a Java version of this.",
            "tokens": [
                50614,
                467,
                390,
                8730,
                294,
                383,
                25472,
                2940,
                2493,
                2057,
                11,
                293,
                257,
                1916,
                295,
                3259,
                2057,
                321,
                611,
                8730,
                257,
                10745,
                3037,
                295,
                341,
                13,
                50914
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19977678908478647,
            "compression_ratio": 1.5560344827586208,
            "no_speech_prob": 0.0149383544921875,
            "words": [
                {
                    "start": 992.69,
                    "end": 993.33,
                    "word": " It",
                    "probability": 0.87060546875
                },
                {
                    "start": 993.33,
                    "end": 993.47,
                    "word": " was",
                    "probability": 0.8994140625
                },
                {
                    "start": 993.47,
                    "end": 993.85,
                    "word": " launched",
                    "probability": 0.84033203125
                },
                {
                    "start": 993.85,
                    "end": 994.25,
                    "word": " in",
                    "probability": 0.9033203125
                },
                {
                    "start": 994.25,
                    "end": 994.73,
                    "word": " C++",
                    "probability": 0.941650390625
                },
                {
                    "start": 994.73,
                    "end": 995.17,
                    "word": " several",
                    "probability": 0.78369140625
                },
                {
                    "start": 995.17,
                    "end": 995.41,
                    "word": " months",
                    "probability": 0.72509765625
                },
                {
                    "start": 995.41,
                    "end": 995.71,
                    "word": " ago",
                    "probability": 0.81787109375
                },
                {
                    "start": 995.71,
                    "end": 996.09,
                    "word": ",",
                    "probability": 0.335693359375
                },
                {
                    "start": 996.09,
                    "end": 996.23,
                    "word": " and",
                    "probability": 0.88818359375
                },
                {
                    "start": 996.47,
                    "end": 996.61,
                    "word": " a couple",
                    "probability": 0.79736328125
                },
                {
                    "start": 996.61,
                    "end": 996.73,
                    "word": " of",
                    "probability": 0.8046875
                },
                {
                    "start": 996.73,
                    "end": 996.91,
                    "word": " weeks",
                    "probability": 0.74755859375
                },
                {
                    "start": 996.91,
                    "end": 997.15,
                    "word": " ago",
                    "probability": 0.81982421875
                },
                {
                    "start": 997.15,
                    "end": 997.33,
                    "word": " we",
                    "probability": 0.595703125
                },
                {
                    "start": 997.33,
                    "end": 997.53,
                    "word": " also",
                    "probability": 0.787109375
                },
                {
                    "start": 997.53,
                    "end": 997.83,
                    "word": " launched",
                    "probability": 0.84033203125
                },
                {
                    "start": 997.97,
                    "end": 998.19,
                    "word": " a Java",
                    "probability": 0.8154296875
                },
                {
                    "start": 998.19,
                    "end": 998.51,
                    "word": " version",
                    "probability": 0.71533203125
                },
                {
                    "start": 998.51,
                    "end": 998.73,
                    "word": " of",
                    "probability": 0.90087890625
                },
                {
                    "start": 998.73,
                    "end": 999.07,
                    "word": " this",
                    "probability": 0.8837890625
                },
                {
                    "start": 999.07,
                    "end": 999.71,
                    "word": ".",
                    "probability": 0.98193359375
                }
            ]
        },
        {
            "id": 140,
            "seek": 101253,
            "start": 999.73,
            "end": 1009.41,
            "text": " I think the analogy we oftentimes use with differential privacy is best compared to cryptography, right?",
            "tokens": [
                50964,
                286,
                519,
                264,
                21663,
                321,
                18349,
                764,
                365,
                15756,
                11427,
                307,
                1151,
                5347,
                281,
                9844,
                5820,
                11,
                558,
                30,
                51414
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19977678908478647,
            "compression_ratio": 1.5560344827586208,
            "no_speech_prob": 0.0149383544921875,
            "words": [
                {
                    "start": 999.73,
                    "end": 999.95,
                    "word": " I",
                    "probability": 0.9931640625
                },
                {
                    "start": 999.95,
                    "end": 1000.15,
                    "word": " think",
                    "probability": 0.8046875
                },
                {
                    "start": 1000.15,
                    "end": 1000.55,
                    "word": " the",
                    "probability": 0.8125
                },
                {
                    "start": 1000.55,
                    "end": 1001.15,
                    "word": " analogy",
                    "probability": 0.73193359375
                },
                {
                    "start": 1001.15,
                    "end": 1001.57,
                    "word": " we",
                    "probability": 0.8974609375
                },
                {
                    "start": 1001.57,
                    "end": 1001.91,
                    "word": " oftentimes",
                    "probability": 0.6357421875
                },
                {
                    "start": 1001.91,
                    "end": 1002.45,
                    "word": " use",
                    "probability": 0.8095703125
                },
                {
                    "start": 1002.45,
                    "end": 1002.83,
                    "word": " with",
                    "probability": 0.802734375
                },
                {
                    "start": 1002.83,
                    "end": 1003.41,
                    "word": " differential",
                    "probability": 0.64404296875
                },
                {
                    "start": 1003.41,
                    "end": 1003.91,
                    "word": " privacy",
                    "probability": 0.8505859375
                },
                {
                    "start": 1003.91,
                    "end": 1004.51,
                    "word": " is",
                    "probability": 0.896484375
                },
                {
                    "start": 1004.51,
                    "end": 1006.87,
                    "word": " best",
                    "probability": 0.76904296875
                },
                {
                    "start": 1006.87,
                    "end": 1007.31,
                    "word": " compared",
                    "probability": 0.8505859375
                },
                {
                    "start": 1007.31,
                    "end": 1007.65,
                    "word": " to",
                    "probability": 0.90380859375
                },
                {
                    "start": 1007.65,
                    "end": 1008.41,
                    "word": " cryptography",
                    "probability": 0.8876953125
                },
                {
                    "start": 1008.41,
                    "end": 1008.65,
                    "word": ",",
                    "probability": 0.57666015625
                },
                {
                    "start": 1008.65,
                    "end": 1008.93,
                    "word": " right",
                    "probability": 0.88671875
                },
                {
                    "start": 1008.93,
                    "end": 1009.41,
                    "word": "?",
                    "probability": 0.84033203125
                }
            ]
        },
        {
            "id": 141,
            "seek": 101253,
            "start": 1009.45,
            "end": 1014.57,
            "text": " There you say, don't roll your own crypto, same thing goes for differential privacy.",
            "tokens": [
                51464,
                821,
                291,
                584,
                11,
                500,
                380,
                3373,
                428,
                1065,
                17240,
                11,
                912,
                551,
                1709,
                337,
                15756,
                11427,
                13,
                51664
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19977678908478647,
            "compression_ratio": 1.5560344827586208,
            "no_speech_prob": 0.0149383544921875,
            "words": [
                {
                    "start": 1009.45,
                    "end": 1010.09,
                    "word": " There",
                    "probability": 0.79296875
                },
                {
                    "start": 1010.09,
                    "end": 1010.27,
                    "word": " you",
                    "probability": 0.7822265625
                },
                {
                    "start": 1010.27,
                    "end": 1010.51,
                    "word": " say",
                    "probability": 0.8828125
                },
                {
                    "start": 1010.51,
                    "end": 1010.81,
                    "word": ",",
                    "probability": 0.396240234375
                },
                {
                    "start": 1010.81,
                    "end": 1010.87,
                    "word": " don't",
                    "probability": 0.7044677734375
                },
                {
                    "start": 1010.87,
                    "end": 1011.07,
                    "word": " roll",
                    "probability": 0.77001953125
                },
                {
                    "start": 1011.07,
                    "end": 1011.23,
                    "word": " your",
                    "probability": 0.8095703125
                },
                {
                    "start": 1011.23,
                    "end": 1011.43,
                    "word": " own",
                    "probability": 0.814453125
                },
                {
                    "start": 1011.43,
                    "end": 1011.75,
                    "word": " crypto",
                    "probability": 0.63916015625
                },
                {
                    "start": 1011.75,
                    "end": 1012.07,
                    "word": ",",
                    "probability": 0.3740234375
                },
                {
                    "start": 1012.07,
                    "end": 1012.53,
                    "word": " same",
                    "probability": 0.78125
                },
                {
                    "start": 1012.53,
                    "end": 1012.79,
                    "word": " thing",
                    "probability": 0.80419921875
                },
                {
                    "start": 1012.79,
                    "end": 1013.27,
                    "word": " goes",
                    "probability": 0.8701171875
                },
                {
                    "start": 1013.27,
                    "end": 1013.55,
                    "word": " for",
                    "probability": 0.89892578125
                },
                {
                    "start": 1013.55,
                    "end": 1013.93,
                    "word": " differential",
                    "probability": 0.71728515625
                },
                {
                    "start": 1013.93,
                    "end": 1014.37,
                    "word": " privacy",
                    "probability": 0.8291015625
                },
                {
                    "start": 1014.37,
                    "end": 1014.57,
                    "word": ".",
                    "probability": 0.89697265625
                }
            ]
        },
        {
            "id": 142,
            "seek": 103977,
            "start": 1014.57,
            "end": 1019.23,
            "text": " There are a lot of gotchas, there's a lot of difficulty involved in rolling it out correctly,",
            "tokens": [
                50364,
                821,
                366,
                257,
                688,
                295,
                658,
                41299,
                11,
                456,
                311,
                257,
                688,
                295,
                10360,
                3288,
                294,
                9439,
                309,
                484,
                8944,
                11,
                50564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18709935846491757,
            "compression_ratio": 1.6761565836298933,
            "no_speech_prob": 0.03900146484375,
            "words": [
                {
                    "start": 1014.57,
                    "end": 1014.67,
                    "word": " There",
                    "probability": 0.458740234375
                },
                {
                    "start": 1014.67,
                    "end": 1014.73,
                    "word": " are",
                    "probability": 0.8505859375
                },
                {
                    "start": 1014.83,
                    "end": 1014.91,
                    "word": " a lot",
                    "probability": 0.8916015625
                },
                {
                    "start": 1014.91,
                    "end": 1015.05,
                    "word": " of",
                    "probability": 0.89501953125
                },
                {
                    "start": 1015.05,
                    "end": 1015.45,
                    "word": " gotchas",
                    "probability": 0.90185546875
                },
                {
                    "start": 1015.45,
                    "end": 1015.65,
                    "word": ",",
                    "probability": 0.517578125
                },
                {
                    "start": 1015.65,
                    "end": 1015.77,
                    "word": " there's",
                    "probability": 0.8740234375
                },
                {
                    "start": 1015.87,
                    "end": 1015.97,
                    "word": " a lot",
                    "probability": 0.884765625
                },
                {
                    "start": 1015.97,
                    "end": 1016.11,
                    "word": " of",
                    "probability": 0.900390625
                },
                {
                    "start": 1016.11,
                    "end": 1016.61,
                    "word": " difficulty",
                    "probability": 0.642578125
                },
                {
                    "start": 1016.61,
                    "end": 1017.35,
                    "word": " involved",
                    "probability": 0.78466796875
                },
                {
                    "start": 1017.35,
                    "end": 1017.57,
                    "word": " in",
                    "probability": 0.89892578125
                },
                {
                    "start": 1017.57,
                    "end": 1017.77,
                    "word": " rolling",
                    "probability": 0.79833984375
                },
                {
                    "start": 1017.77,
                    "end": 1017.93,
                    "word": " it",
                    "probability": 0.90234375
                },
                {
                    "start": 1017.93,
                    "end": 1018.13,
                    "word": " out",
                    "probability": 0.8154296875
                },
                {
                    "start": 1018.13,
                    "end": 1018.85,
                    "word": " correctly",
                    "probability": 0.3369140625
                },
                {
                    "start": 1018.85,
                    "end": 1019.23,
                    "word": ",",
                    "probability": 0.306884765625
                }
            ]
        },
        {
            "id": 143,
            "seek": 103977,
            "start": 1019.23,
            "end": 1025.05,
            "text": " and an incorrect or partial implementation of DP is in some cases worse than none at all,",
            "tokens": [
                50614,
                293,
                364,
                18424,
                420,
                14641,
                11420,
                295,
                42796,
                307,
                294,
                512,
                3331,
                5324,
                813,
                6022,
                412,
                439,
                11,
                50864
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18709935846491757,
            "compression_ratio": 1.6761565836298933,
            "no_speech_prob": 0.03900146484375,
            "words": [
                {
                    "start": 1019.23,
                    "end": 1019.71,
                    "word": " and",
                    "probability": 0.88232421875
                },
                {
                    "start": 1019.71,
                    "end": 1019.95,
                    "word": " an",
                    "probability": 0.63330078125
                },
                {
                    "start": 1019.95,
                    "end": 1020.45,
                    "word": " incorrect",
                    "probability": 0.947265625
                },
                {
                    "start": 1020.45,
                    "end": 1021.11,
                    "word": " or",
                    "probability": 0.814453125
                },
                {
                    "start": 1021.11,
                    "end": 1021.59,
                    "word": " partial",
                    "probability": 0.8681640625
                },
                {
                    "start": 1021.59,
                    "end": 1022.39,
                    "word": " implementation",
                    "probability": 0.72509765625
                },
                {
                    "start": 1022.39,
                    "end": 1022.71,
                    "word": " of",
                    "probability": 0.904296875
                },
                {
                    "start": 1022.71,
                    "end": 1022.97,
                    "word": " DP",
                    "probability": 0.7578125
                },
                {
                    "start": 1022.97,
                    "end": 1023.33,
                    "word": " is",
                    "probability": 0.89599609375
                },
                {
                    "start": 1023.33,
                    "end": 1023.49,
                    "word": " in",
                    "probability": 0.68701171875
                },
                {
                    "start": 1023.49,
                    "end": 1023.65,
                    "word": " some",
                    "probability": 0.8212890625
                },
                {
                    "start": 1023.65,
                    "end": 1023.95,
                    "word": " cases",
                    "probability": 0.8193359375
                },
                {
                    "start": 1023.95,
                    "end": 1024.29,
                    "word": " worse",
                    "probability": 0.8203125
                },
                {
                    "start": 1024.29,
                    "end": 1024.53,
                    "word": " than",
                    "probability": 0.8857421875
                },
                {
                    "start": 1024.53,
                    "end": 1024.67,
                    "word": " none",
                    "probability": 0.82568359375
                },
                {
                    "start": 1024.67,
                    "end": 1024.79,
                    "word": " at",
                    "probability": 0.89404296875
                },
                {
                    "start": 1024.79,
                    "end": 1024.99,
                    "word": " all",
                    "probability": 0.896484375
                },
                {
                    "start": 1024.99,
                    "end": 1025.05,
                    "word": ",",
                    "probability": 0.3525390625
                }
            ]
        },
        {
            "id": 144,
            "seek": 103977,
            "start": 1025.05,
            "end": 1028.83,
            "text": " because you might have a gaping hole in your system that you're not aware of.",
            "tokens": [
                50914,
                570,
                291,
                1062,
                362,
                257,
                7417,
                278,
                5458,
                294,
                428,
                1185,
                300,
                291,
                434,
                406,
                3650,
                295,
                13,
                51064
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18709935846491757,
            "compression_ratio": 1.6761565836298933,
            "no_speech_prob": 0.03900146484375,
            "words": [
                {
                    "start": 1025.05,
                    "end": 1025.25,
                    "word": " because",
                    "probability": 0.83203125
                },
                {
                    "start": 1025.25,
                    "end": 1025.39,
                    "word": " you",
                    "probability": 0.89599609375
                },
                {
                    "start": 1025.39,
                    "end": 1025.53,
                    "word": " might",
                    "probability": 0.81103515625
                },
                {
                    "start": 1025.53,
                    "end": 1025.67,
                    "word": " have",
                    "probability": 0.8798828125
                },
                {
                    "start": 1025.79,
                    "end": 1026.07,
                    "word": " a gaping",
                    "probability": 0.92919921875
                },
                {
                    "start": 1026.07,
                    "end": 1026.31,
                    "word": " hole",
                    "probability": 0.794921875
                },
                {
                    "start": 1026.31,
                    "end": 1026.45,
                    "word": " in",
                    "probability": 0.90234375
                },
                {
                    "start": 1026.45,
                    "end": 1026.57,
                    "word": " your",
                    "probability": 0.81103515625
                },
                {
                    "start": 1026.57,
                    "end": 1026.95,
                    "word": " system",
                    "probability": 0.86767578125
                },
                {
                    "start": 1026.95,
                    "end": 1027.93,
                    "word": " that",
                    "probability": 0.841796875
                },
                {
                    "start": 1027.93,
                    "end": 1028.05,
                    "word": " you're",
                    "probability": 0.91845703125
                },
                {
                    "start": 1028.05,
                    "end": 1028.21,
                    "word": " not",
                    "probability": 0.8994140625
                },
                {
                    "start": 1028.21,
                    "end": 1028.41,
                    "word": " aware",
                    "probability": 0.83203125
                },
                {
                    "start": 1028.41,
                    "end": 1028.67,
                    "word": " of",
                    "probability": 0.90087890625
                },
                {
                    "start": 1028.67,
                    "end": 1028.83,
                    "word": ".",
                    "probability": 0.92431640625
                }
            ]
        },
        {
            "id": 145,
            "seek": 103977,
            "start": 1028.83,
            "end": 1031.99,
            "text": " So reusing a hardened implementation is just a good idea.",
            "tokens": [
                51114,
                407,
                319,
                7981,
                257,
                42605,
                11420,
                307,
                445,
                257,
                665,
                1558,
                13,
                51264
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18709935846491757,
            "compression_ratio": 1.6761565836298933,
            "no_speech_prob": 0.03900146484375,
            "words": [
                {
                    "start": 1028.83,
                    "end": 1028.95,
                    "word": " So",
                    "probability": 0.7177734375
                },
                {
                    "start": 1028.95,
                    "end": 1029.41,
                    "word": " reusing",
                    "probability": 0.812255859375
                },
                {
                    "start": 1029.65,
                    "end": 1029.89,
                    "word": " a hardened",
                    "probability": 0.79833984375
                },
                {
                    "start": 1029.89,
                    "end": 1030.33,
                    "word": " implementation",
                    "probability": 0.7265625
                },
                {
                    "start": 1030.33,
                    "end": 1030.73,
                    "word": " is",
                    "probability": 0.90185546875
                },
                {
                    "start": 1030.73,
                    "end": 1030.91,
                    "word": " just",
                    "probability": 0.87890625
                },
                {
                    "start": 1031.07,
                    "end": 1031.23,
                    "word": " a good",
                    "probability": 0.87890625
                },
                {
                    "start": 1031.23,
                    "end": 1031.59,
                    "word": " idea",
                    "probability": 0.77685546875
                },
                {
                    "start": 1031.59,
                    "end": 1031.99,
                    "word": ".",
                    "probability": 0.99755859375
                }
            ]
        },
        {
            "id": 146,
            "seek": 103977,
            "start": 1033.37,
            "end": 1037.29,
            "text": " So the second step we did was to build a SQL engine,",
            "tokens": [
                51314,
                407,
                264,
                1150,
                1823,
                321,
                630,
                390,
                281,
                1322,
                257,
                19200,
                2848,
                11,
                51464
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18709935846491757,
            "compression_ratio": 1.6761565836298933,
            "no_speech_prob": 0.03900146484375,
            "words": [
                {
                    "start": 1033.37,
                    "end": 1033.77,
                    "word": " So",
                    "probability": 0.61474609375
                },
                {
                    "start": 1033.77,
                    "end": 1033.91,
                    "word": " the",
                    "probability": 0.77783203125
                },
                {
                    "start": 1033.91,
                    "end": 1034.15,
                    "word": " second",
                    "probability": 0.79443359375
                },
                {
                    "start": 1034.15,
                    "end": 1034.49,
                    "word": " step",
                    "probability": 0.80810546875
                },
                {
                    "start": 1034.49,
                    "end": 1034.65,
                    "word": " we",
                    "probability": 0.89404296875
                },
                {
                    "start": 1034.65,
                    "end": 1034.81,
                    "word": " did",
                    "probability": 0.9013671875
                },
                {
                    "start": 1034.81,
                    "end": 1034.99,
                    "word": " was",
                    "probability": 0.88330078125
                },
                {
                    "start": 1034.99,
                    "end": 1035.19,
                    "word": " to",
                    "probability": 0.90234375
                },
                {
                    "start": 1035.19,
                    "end": 1035.75,
                    "word": " build",
                    "probability": 0.86328125
                },
                {
                    "start": 1036.37,
                    "end": 1036.63,
                    "word": " a SQL",
                    "probability": 0.873046875
                },
                {
                    "start": 1036.63,
                    "end": 1037.03,
                    "word": " engine",
                    "probability": 0.66552734375
                },
                {
                    "start": 1037.03,
                    "end": 1037.29,
                    "word": ",",
                    "probability": 0.376708984375
                }
            ]
        },
        {
            "id": 147,
            "seek": 103977,
            "start": 1037.29,
            "end": 1041.81,
            "text": " and there were lots of languages we could have used, lots of ways we could have gotten about this,",
            "tokens": [
                51514,
                293,
                456,
                645,
                3195,
                295,
                8650,
                321,
                727,
                362,
                1143,
                11,
                3195,
                295,
                2098,
                321,
                727,
                362,
                5768,
                466,
                341,
                11,
                51714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18709935846491757,
            "compression_ratio": 1.6761565836298933,
            "no_speech_prob": 0.03900146484375,
            "words": [
                {
                    "start": 1037.29,
                    "end": 1037.59,
                    "word": " and",
                    "probability": 0.89404296875
                },
                {
                    "start": 1037.59,
                    "end": 1037.73,
                    "word": " there",
                    "probability": 0.85107421875
                },
                {
                    "start": 1037.73,
                    "end": 1037.85,
                    "word": " were",
                    "probability": 0.8837890625
                },
                {
                    "start": 1037.85,
                    "end": 1038.07,
                    "word": " lots",
                    "probability": 0.79296875
                },
                {
                    "start": 1038.07,
                    "end": 1038.19,
                    "word": " of",
                    "probability": 0.904296875
                },
                {
                    "start": 1038.19,
                    "end": 1038.51,
                    "word": " languages",
                    "probability": 0.85107421875
                },
                {
                    "start": 1038.51,
                    "end": 1038.73,
                    "word": " we",
                    "probability": 0.8916015625
                },
                {
                    "start": 1038.73,
                    "end": 1038.85,
                    "word": " could",
                    "probability": 0.8203125
                },
                {
                    "start": 1038.85,
                    "end": 1038.97,
                    "word": " have",
                    "probability": 0.8115234375
                },
                {
                    "start": 1038.97,
                    "end": 1039.23,
                    "word": " used",
                    "probability": 0.81689453125
                },
                {
                    "start": 1039.23,
                    "end": 1039.33,
                    "word": ",",
                    "probability": 0.92822265625
                },
                {
                    "start": 1039.33,
                    "end": 1039.49,
                    "word": " lots",
                    "probability": 0.79296875
                },
                {
                    "start": 1039.49,
                    "end": 1039.61,
                    "word": " of",
                    "probability": 0.90185546875
                },
                {
                    "start": 1039.61,
                    "end": 1039.83,
                    "word": " ways",
                    "probability": 0.81689453125
                },
                {
                    "start": 1039.83,
                    "end": 1039.97,
                    "word": " we",
                    "probability": 0.900390625
                },
                {
                    "start": 1039.97,
                    "end": 1040.09,
                    "word": " could",
                    "probability": 0.82421875
                },
                {
                    "start": 1040.09,
                    "end": 1040.19,
                    "word": " have",
                    "probability": 0.8583984375
                },
                {
                    "start": 1040.19,
                    "end": 1040.47,
                    "word": " gotten",
                    "probability": 0.71435546875
                },
                {
                    "start": 1040.47,
                    "end": 1040.99,
                    "word": " about",
                    "probability": 0.82080078125
                },
                {
                    "start": 1040.99,
                    "end": 1041.39,
                    "word": " this",
                    "probability": 0.88037109375
                },
                {
                    "start": 1041.39,
                    "end": 1041.81,
                    "word": ",",
                    "probability": 0.74853515625
                }
            ]
        },
        {
            "id": 148,
            "seek": 106867,
            "start": 1041.81,
            "end": 1044.75,
            "text": " but SQL really made the most sense, right?",
            "tokens": [
                50364,
                457,
                19200,
                534,
                1027,
                264,
                881,
                2020,
                11,
                558,
                30,
                50464
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18655710821521693,
            "compression_ratio": 1.605072463768116,
            "no_speech_prob": 0.00638580322265625,
            "words": [
                {
                    "start": 1041.81,
                    "end": 1042.29,
                    "word": " but",
                    "probability": 0.07952880859375
                },
                {
                    "start": 1042.29,
                    "end": 1042.65,
                    "word": " SQL",
                    "probability": 0.77734375
                },
                {
                    "start": 1042.65,
                    "end": 1042.95,
                    "word": " really",
                    "probability": 0.76611328125
                },
                {
                    "start": 1042.95,
                    "end": 1043.43,
                    "word": " made",
                    "probability": 0.81640625
                },
                {
                    "start": 1043.43,
                    "end": 1043.61,
                    "word": " the",
                    "probability": 0.8154296875
                },
                {
                    "start": 1043.61,
                    "end": 1043.87,
                    "word": " most",
                    "probability": 0.81201171875
                },
                {
                    "start": 1043.87,
                    "end": 1044.35,
                    "word": " sense",
                    "probability": 0.7529296875
                },
                {
                    "start": 1044.35,
                    "end": 1044.47,
                    "word": ",",
                    "probability": 0.60400390625
                },
                {
                    "start": 1044.47,
                    "end": 1044.67,
                    "word": " right",
                    "probability": 0.84619140625
                },
                {
                    "start": 1044.67,
                    "end": 1044.75,
                    "word": "?",
                    "probability": 0.78955078125
                }
            ]
        },
        {
            "id": 149,
            "seek": 106867,
            "start": 1044.75,
            "end": 1049.03,
            "text": " It's the language that most developers are familiar with,",
            "tokens": [
                50514,
                467,
                311,
                264,
                2856,
                300,
                881,
                8849,
                366,
                4963,
                365,
                11,
                50664
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18655710821521693,
            "compression_ratio": 1.605072463768116,
            "no_speech_prob": 0.00638580322265625,
            "words": [
                {
                    "start": 1044.75,
                    "end": 1045.01,
                    "word": " It's",
                    "probability": 0.937744140625
                },
                {
                    "start": 1045.01,
                    "end": 1045.51,
                    "word": " the",
                    "probability": 0.81396484375
                },
                {
                    "start": 1045.51,
                    "end": 1046.71,
                    "word": " language",
                    "probability": 0.55322265625
                },
                {
                    "start": 1046.71,
                    "end": 1047.09,
                    "word": " that",
                    "probability": 0.87451171875
                },
                {
                    "start": 1047.09,
                    "end": 1047.41,
                    "word": " most",
                    "probability": 0.802734375
                },
                {
                    "start": 1047.41,
                    "end": 1047.81,
                    "word": " developers",
                    "probability": 0.703125
                },
                {
                    "start": 1047.81,
                    "end": 1048.01,
                    "word": " are",
                    "probability": 0.8984375
                },
                {
                    "start": 1048.01,
                    "end": 1048.29,
                    "word": " familiar",
                    "probability": 0.7626953125
                },
                {
                    "start": 1048.29,
                    "end": 1048.65,
                    "word": " with",
                    "probability": 0.81640625
                },
                {
                    "start": 1048.65,
                    "end": 1049.03,
                    "word": ",",
                    "probability": 0.2177734375
                }
            ]
        },
        {
            "id": 150,
            "seek": 106867,
            "start": 1049.03,
            "end": 1054.39,
            "text": " you don't have to be an expert to get in and start doing an analysis on data sets,",
            "tokens": [
                50714,
                291,
                500,
                380,
                362,
                281,
                312,
                364,
                5844,
                281,
                483,
                294,
                293,
                722,
                884,
                364,
                5215,
                322,
                1412,
                6352,
                11,
                50964
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18655710821521693,
            "compression_ratio": 1.605072463768116,
            "no_speech_prob": 0.00638580322265625,
            "words": [
                {
                    "start": 1049.03,
                    "end": 1049.31,
                    "word": " you",
                    "probability": 0.85595703125
                },
                {
                    "start": 1049.31,
                    "end": 1049.43,
                    "word": " don't",
                    "probability": 0.94580078125
                },
                {
                    "start": 1049.43,
                    "end": 1049.59,
                    "word": " have",
                    "probability": 0.88330078125
                },
                {
                    "start": 1049.59,
                    "end": 1049.69,
                    "word": " to",
                    "probability": 0.90673828125
                },
                {
                    "start": 1049.69,
                    "end": 1049.79,
                    "word": " be",
                    "probability": 0.90673828125
                },
                {
                    "start": 1049.79,
                    "end": 1049.95,
                    "word": " an",
                    "probability": 0.90869140625
                },
                {
                    "start": 1049.95,
                    "end": 1050.21,
                    "word": " expert",
                    "probability": 0.802734375
                },
                {
                    "start": 1050.21,
                    "end": 1050.45,
                    "word": " to",
                    "probability": 0.90283203125
                },
                {
                    "start": 1050.45,
                    "end": 1050.61,
                    "word": " get",
                    "probability": 0.888671875
                },
                {
                    "start": 1050.61,
                    "end": 1050.77,
                    "word": " in",
                    "probability": 0.908203125
                },
                {
                    "start": 1050.77,
                    "end": 1050.89,
                    "word": " and",
                    "probability": 0.89208984375
                },
                {
                    "start": 1050.89,
                    "end": 1051.17,
                    "word": " start",
                    "probability": 0.87939453125
                },
                {
                    "start": 1051.17,
                    "end": 1051.57,
                    "word": " doing",
                    "probability": 0.90380859375
                },
                {
                    "start": 1051.57,
                    "end": 1051.77,
                    "word": " an",
                    "probability": 0.8935546875
                },
                {
                    "start": 1051.77,
                    "end": 1052.27,
                    "word": " analysis",
                    "probability": 0.76611328125
                },
                {
                    "start": 1052.27,
                    "end": 1053.15,
                    "word": " on",
                    "probability": 0.90576171875
                },
                {
                    "start": 1053.15,
                    "end": 1053.37,
                    "word": " data",
                    "probability": 0.425048828125
                },
                {
                    "start": 1053.37,
                    "end": 1053.81,
                    "word": " sets",
                    "probability": 0.79638671875
                },
                {
                    "start": 1053.81,
                    "end": 1054.39,
                    "word": ",",
                    "probability": 0.384521484375
                }
            ]
        },
        {
            "id": 151,
            "seek": 106867,
            "start": 1054.93,
            "end": 1057.67,
            "text": " and so it just felt like the right choice.",
            "tokens": [
                51014,
                293,
                370,
                309,
                445,
                2762,
                411,
                264,
                558,
                3922,
                13,
                51114
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18655710821521693,
            "compression_ratio": 1.605072463768116,
            "no_speech_prob": 0.00638580322265625,
            "words": [
                {
                    "start": 1054.93,
                    "end": 1055.41,
                    "word": " and",
                    "probability": 0.826171875
                },
                {
                    "start": 1055.41,
                    "end": 1055.59,
                    "word": " so",
                    "probability": 0.89208984375
                },
                {
                    "start": 1055.59,
                    "end": 1055.73,
                    "word": " it",
                    "probability": 0.86962890625
                },
                {
                    "start": 1055.73,
                    "end": 1055.87,
                    "word": " just",
                    "probability": 0.88037109375
                },
                {
                    "start": 1055.87,
                    "end": 1056.05,
                    "word": " felt",
                    "probability": 0.8115234375
                },
                {
                    "start": 1056.05,
                    "end": 1056.23,
                    "word": " like",
                    "probability": 0.8798828125
                },
                {
                    "start": 1056.23,
                    "end": 1056.35,
                    "word": " the",
                    "probability": 0.8134765625
                },
                {
                    "start": 1056.35,
                    "end": 1056.61,
                    "word": " right",
                    "probability": 0.880859375
                },
                {
                    "start": 1056.61,
                    "end": 1057.21,
                    "word": " choice",
                    "probability": 0.79736328125
                },
                {
                    "start": 1057.21,
                    "end": 1057.67,
                    "word": ".",
                    "probability": 0.94384765625
                }
            ]
        },
        {
            "id": 152,
            "seek": 106867,
            "start": 1057.67,
            "end": 1063.47,
            "text": " So I'm just going to give a little bit of an overview of some of the challenges involved in this.",
            "tokens": [
                51164,
                407,
                286,
                478,
                445,
                516,
                281,
                976,
                257,
                707,
                857,
                295,
                364,
                12492,
                295,
                512,
                295,
                264,
                4759,
                3288,
                294,
                341,
                13,
                51414
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18655710821521693,
            "compression_ratio": 1.605072463768116,
            "no_speech_prob": 0.00638580322265625,
            "words": [
                {
                    "start": 1057.67,
                    "end": 1058.13,
                    "word": " So",
                    "probability": 0.81103515625
                },
                {
                    "start": 1058.13,
                    "end": 1058.25,
                    "word": " I'm",
                    "probability": 0.90869140625
                },
                {
                    "start": 1058.25,
                    "end": 1058.39,
                    "word": " just",
                    "probability": 0.86962890625
                },
                {
                    "start": 1058.39,
                    "end": 1058.53,
                    "word": " going",
                    "probability": 0.68994140625
                },
                {
                    "start": 1058.53,
                    "end": 1058.59,
                    "word": " to",
                    "probability": 0.90478515625
                },
                {
                    "start": 1058.59,
                    "end": 1058.73,
                    "word": " give",
                    "probability": 0.80078125
                },
                {
                    "start": 1058.95,
                    "end": 1059.13,
                    "word": " a little",
                    "probability": 0.73974609375
                },
                {
                    "start": 1059.13,
                    "end": 1059.31,
                    "word": " bit",
                    "probability": 0.89208984375
                },
                {
                    "start": 1059.31,
                    "end": 1059.39,
                    "word": " of",
                    "probability": 0.908203125
                },
                {
                    "start": 1059.39,
                    "end": 1059.49,
                    "word": " an",
                    "probability": 0.9052734375
                },
                {
                    "start": 1059.49,
                    "end": 1059.79,
                    "word": " overview",
                    "probability": 0.822265625
                },
                {
                    "start": 1059.79,
                    "end": 1060.01,
                    "word": " of",
                    "probability": 0.89794921875
                },
                {
                    "start": 1060.01,
                    "end": 1060.87,
                    "word": " some",
                    "probability": 0.350341796875
                },
                {
                    "start": 1060.87,
                    "end": 1061.55,
                    "word": " of",
                    "probability": 0.90576171875
                },
                {
                    "start": 1061.55,
                    "end": 1061.63,
                    "word": " the",
                    "probability": 0.8134765625
                },
                {
                    "start": 1061.63,
                    "end": 1061.99,
                    "word": " challenges",
                    "probability": 0.74072265625
                },
                {
                    "start": 1061.99,
                    "end": 1062.51,
                    "word": " involved",
                    "probability": 0.7607421875
                },
                {
                    "start": 1062.51,
                    "end": 1062.69,
                    "word": " in",
                    "probability": 0.89990234375
                },
                {
                    "start": 1062.69,
                    "end": 1062.99,
                    "word": " this",
                    "probability": 0.8837890625
                },
                {
                    "start": 1062.99,
                    "end": 1063.47,
                    "word": ".",
                    "probability": 0.984375
                }
            ]
        },
        {
            "id": 153,
            "seek": 106867,
            "start": 1063.81,
            "end": 1068.17,
            "text": " So one of the big ones is that SQL allows pretty much anything.",
            "tokens": [
                51464,
                407,
                472,
                295,
                264,
                955,
                2306,
                307,
                300,
                19200,
                4045,
                1238,
                709,
                1340,
                13,
                51614
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18655710821521693,
            "compression_ratio": 1.605072463768116,
            "no_speech_prob": 0.00638580322265625,
            "words": [
                {
                    "start": 1063.81,
                    "end": 1064.23,
                    "word": " So",
                    "probability": 0.74609375
                },
                {
                    "start": 1064.23,
                    "end": 1064.75,
                    "word": " one",
                    "probability": 0.666015625
                },
                {
                    "start": 1064.75,
                    "end": 1064.83,
                    "word": " of",
                    "probability": 0.90380859375
                },
                {
                    "start": 1064.83,
                    "end": 1064.91,
                    "word": " the",
                    "probability": 0.8134765625
                },
                {
                    "start": 1064.91,
                    "end": 1065.09,
                    "word": " big",
                    "probability": 0.896484375
                },
                {
                    "start": 1065.09,
                    "end": 1065.31,
                    "word": " ones",
                    "probability": 0.8203125
                },
                {
                    "start": 1065.31,
                    "end": 1065.49,
                    "word": " is",
                    "probability": 0.90869140625
                },
                {
                    "start": 1065.49,
                    "end": 1065.67,
                    "word": " that",
                    "probability": 0.87646484375
                },
                {
                    "start": 1065.67,
                    "end": 1065.95,
                    "word": " SQL",
                    "probability": 0.912109375
                },
                {
                    "start": 1065.95,
                    "end": 1066.47,
                    "word": " allows",
                    "probability": 0.775390625
                },
                {
                    "start": 1066.47,
                    "end": 1066.95,
                    "word": " pretty",
                    "probability": 0.86376953125
                },
                {
                    "start": 1066.95,
                    "end": 1067.23,
                    "word": " much",
                    "probability": 0.82373046875
                },
                {
                    "start": 1067.23,
                    "end": 1067.69,
                    "word": " anything",
                    "probability": 0.83935546875
                },
                {
                    "start": 1067.69,
                    "end": 1068.17,
                    "word": ".",
                    "probability": 0.720703125
                }
            ]
        },
        {
            "id": 154,
            "seek": 106867,
            "start": 1068.17,
            "end": 1070.71,
            "text": " A lot of implementations are actually Turing-complete,",
            "tokens": [
                51664,
                316,
                688,
                295,
                4445,
                763,
                366,
                767,
                314,
                1345,
                12,
                1112,
                17220,
                11,
                51764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18655710821521693,
            "compression_ratio": 1.605072463768116,
            "no_speech_prob": 0.00638580322265625,
            "words": [
                {
                    "start": 1068.17,
                    "end": 1068.41,
                    "word": " A",
                    "probability": 0.99365234375
                },
                {
                    "start": 1068.41,
                    "end": 1068.55,
                    "word": " lot",
                    "probability": 0.8876953125
                },
                {
                    "start": 1068.55,
                    "end": 1068.67,
                    "word": " of",
                    "probability": 0.90625
                },
                {
                    "start": 1068.67,
                    "end": 1069.21,
                    "word": " implementations",
                    "probability": 0.9140625
                },
                {
                    "start": 1069.21,
                    "end": 1069.37,
                    "word": " are",
                    "probability": 0.8984375
                },
                {
                    "start": 1069.37,
                    "end": 1069.61,
                    "word": " actually",
                    "probability": 0.814453125
                },
                {
                    "start": 1069.61,
                    "end": 1069.83,
                    "word": " Turing",
                    "probability": 0.9091796875
                },
                {
                    "start": 1069.83,
                    "end": 1070.21,
                    "word": "-complete",
                    "probability": 0.7565104166666666
                },
                {
                    "start": 1070.21,
                    "end": 1070.71,
                    "word": ",",
                    "probability": 0.64794921875
                }
            ]
        },
        {
            "id": 155,
            "seek": 109867,
            "start": 1070.71,
            "end": 1072.95,
            "text": " so enforcing differential privacy,",
            "tokens": [
                50364,
                370,
                25495,
                2175,
                15756,
                11427,
                11,
                50464
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20032428697032748,
            "compression_ratio": 1.6738351254480286,
            "no_speech_prob": 0.0009255409240722656,
            "words": [
                {
                    "start": 1070.71,
                    "end": 1070.95,
                    "word": " so",
                    "probability": 0.038909912109375
                },
                {
                    "start": 1070.95,
                    "end": 1071.63,
                    "word": " enforcing",
                    "probability": 0.89599609375
                },
                {
                    "start": 1071.63,
                    "end": 1072.15,
                    "word": " differential",
                    "probability": 0.71435546875
                },
                {
                    "start": 1072.15,
                    "end": 1072.65,
                    "word": " privacy",
                    "probability": 0.8505859375
                },
                {
                    "start": 1072.65,
                    "end": 1072.95,
                    "word": ",",
                    "probability": 0.1776123046875
                }
            ]
        },
        {
            "id": 156,
            "seek": 109867,
            "start": 1073.25,
            "end": 1079.59,
            "text": " while at the same time allowing non-experts to write SQL that they're basically familiar with is a big challenge.",
            "tokens": [
                50514,
                1339,
                412,
                264,
                912,
                565,
                8293,
                2107,
                12,
                3121,
                610,
                1373,
                281,
                2464,
                19200,
                300,
                436,
                434,
                1936,
                4963,
                365,
                307,
                257,
                955,
                3430,
                13,
                50764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20032428697032748,
            "compression_ratio": 1.6738351254480286,
            "no_speech_prob": 0.0009255409240722656,
            "words": [
                {
                    "start": 1073.25,
                    "end": 1073.59,
                    "word": " while",
                    "probability": 0.9208984375
                },
                {
                    "start": 1073.59,
                    "end": 1073.73,
                    "word": " at",
                    "probability": 0.900390625
                },
                {
                    "start": 1073.73,
                    "end": 1073.89,
                    "word": " the",
                    "probability": 0.8115234375
                },
                {
                    "start": 1073.89,
                    "end": 1074.17,
                    "word": " same",
                    "probability": 0.82421875
                },
                {
                    "start": 1074.17,
                    "end": 1074.53,
                    "word": " time",
                    "probability": 0.818359375
                },
                {
                    "start": 1074.53,
                    "end": 1074.83,
                    "word": " allowing",
                    "probability": 0.7021484375
                },
                {
                    "start": 1074.83,
                    "end": 1075.09,
                    "word": " non",
                    "probability": 0.88037109375
                },
                {
                    "start": 1075.09,
                    "end": 1075.51,
                    "word": "-experts",
                    "probability": 0.86474609375
                },
                {
                    "start": 1075.51,
                    "end": 1075.65,
                    "word": " to",
                    "probability": 0.83251953125
                },
                {
                    "start": 1075.65,
                    "end": 1075.81,
                    "word": " write",
                    "probability": 0.8349609375
                },
                {
                    "start": 1075.81,
                    "end": 1076.11,
                    "word": " SQL",
                    "probability": 0.8369140625
                },
                {
                    "start": 1076.11,
                    "end": 1076.29,
                    "word": " that",
                    "probability": 0.85791015625
                },
                {
                    "start": 1076.29,
                    "end": 1076.51,
                    "word": " they're",
                    "probability": 0.819091796875
                },
                {
                    "start": 1076.51,
                    "end": 1076.93,
                    "word": " basically",
                    "probability": 0.853515625
                },
                {
                    "start": 1076.93,
                    "end": 1077.37,
                    "word": " familiar",
                    "probability": 0.75390625
                },
                {
                    "start": 1077.37,
                    "end": 1077.65,
                    "word": " with",
                    "probability": 0.81591796875
                },
                {
                    "start": 1077.65,
                    "end": 1077.91,
                    "word": " is",
                    "probability": 0.4775390625
                },
                {
                    "start": 1078.27,
                    "end": 1078.43,
                    "word": " a big",
                    "probability": 0.8955078125
                },
                {
                    "start": 1078.43,
                    "end": 1078.99,
                    "word": " challenge",
                    "probability": 0.78857421875
                },
                {
                    "start": 1078.99,
                    "end": 1079.59,
                    "word": ".",
                    "probability": 0.765625
                }
            ]
        },
        {
            "id": 157,
            "seek": 109867,
            "start": 1079.71,
            "end": 1084.65,
            "text": " A lot of challenges come up around things like joins between different data sets,",
            "tokens": [
                50814,
                316,
                688,
                295,
                4759,
                808,
                493,
                926,
                721,
                411,
                24397,
                1296,
                819,
                1412,
                6352,
                11,
                51014
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20032428697032748,
            "compression_ratio": 1.6738351254480286,
            "no_speech_prob": 0.0009255409240722656,
            "words": [
                {
                    "start": 1079.71,
                    "end": 1080.31,
                    "word": " A",
                    "probability": 0.8896484375
                },
                {
                    "start": 1080.31,
                    "end": 1080.63,
                    "word": " lot",
                    "probability": 0.89208984375
                },
                {
                    "start": 1080.63,
                    "end": 1080.75,
                    "word": " of",
                    "probability": 0.90673828125
                },
                {
                    "start": 1080.75,
                    "end": 1081.11,
                    "word": " challenges",
                    "probability": 0.74951171875
                },
                {
                    "start": 1081.11,
                    "end": 1081.47,
                    "word": " come",
                    "probability": 0.81005859375
                },
                {
                    "start": 1081.47,
                    "end": 1081.65,
                    "word": " up",
                    "probability": 0.9150390625
                },
                {
                    "start": 1081.65,
                    "end": 1081.95,
                    "word": " around",
                    "probability": 0.8740234375
                },
                {
                    "start": 1081.95,
                    "end": 1082.33,
                    "word": " things",
                    "probability": 0.740234375
                },
                {
                    "start": 1082.33,
                    "end": 1082.69,
                    "word": " like",
                    "probability": 0.875
                },
                {
                    "start": 1082.69,
                    "end": 1083.19,
                    "word": " joins",
                    "probability": 0.9189453125
                },
                {
                    "start": 1083.19,
                    "end": 1083.57,
                    "word": " between",
                    "probability": 0.7646484375
                },
                {
                    "start": 1083.57,
                    "end": 1083.91,
                    "word": " different",
                    "probability": 0.69287109375
                },
                {
                    "start": 1083.91,
                    "end": 1084.09,
                    "word": " data",
                    "probability": 0.483642578125
                },
                {
                    "start": 1084.09,
                    "end": 1084.51,
                    "word": " sets",
                    "probability": 0.80126953125
                },
                {
                    "start": 1084.51,
                    "end": 1084.65,
                    "word": ",",
                    "probability": 0.2049560546875
                }
            ]
        },
        {
            "id": 158,
            "seek": 109867,
            "start": 1084.65,
            "end": 1088.57,
            "text": " or cardinality changes, if you're allowed to just multiply rows endlessly,",
            "tokens": [
                51064,
                420,
                2920,
                259,
                1860,
                2962,
                11,
                498,
                291,
                434,
                4350,
                281,
                445,
                12972,
                13241,
                44920,
                11,
                51214
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20032428697032748,
            "compression_ratio": 1.6738351254480286,
            "no_speech_prob": 0.0009255409240722656,
            "words": [
                {
                    "start": 1084.65,
                    "end": 1084.71,
                    "word": " or",
                    "probability": 0.89453125
                },
                {
                    "start": 1084.71,
                    "end": 1085.35,
                    "word": " cardinality",
                    "probability": 0.8064778645833334
                },
                {
                    "start": 1085.35,
                    "end": 1085.83,
                    "word": " changes",
                    "probability": 0.8076171875
                },
                {
                    "start": 1085.83,
                    "end": 1085.95,
                    "word": ",",
                    "probability": 0.284423828125
                },
                {
                    "start": 1085.95,
                    "end": 1086.17,
                    "word": " if",
                    "probability": 0.88720703125
                },
                {
                    "start": 1086.17,
                    "end": 1086.29,
                    "word": " you're",
                    "probability": 0.9189453125
                },
                {
                    "start": 1086.29,
                    "end": 1086.47,
                    "word": " allowed",
                    "probability": 0.79833984375
                },
                {
                    "start": 1086.47,
                    "end": 1086.61,
                    "word": " to",
                    "probability": 0.90771484375
                },
                {
                    "start": 1086.61,
                    "end": 1086.81,
                    "word": " just",
                    "probability": 0.8798828125
                },
                {
                    "start": 1086.81,
                    "end": 1087.17,
                    "word": " multiply",
                    "probability": 0.80322265625
                },
                {
                    "start": 1087.17,
                    "end": 1087.55,
                    "word": " rows",
                    "probability": 0.81005859375
                },
                {
                    "start": 1087.55,
                    "end": 1088.01,
                    "word": " endlessly",
                    "probability": 0.52490234375
                },
                {
                    "start": 1088.01,
                    "end": 1088.57,
                    "word": ",",
                    "probability": 0.57861328125
                }
            ]
        },
        {
            "id": 159,
            "seek": 109867,
            "start": 1088.57,
            "end": 1092.35,
            "text": " lots of privacy implications come out of that.",
            "tokens": [
                51264,
                3195,
                295,
                11427,
                16602,
                808,
                484,
                295,
                300,
                13,
                51414
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20032428697032748,
            "compression_ratio": 1.6738351254480286,
            "no_speech_prob": 0.0009255409240722656,
            "words": [
                {
                    "start": 1088.57,
                    "end": 1089.21,
                    "word": " lots",
                    "probability": 0.75927734375
                },
                {
                    "start": 1089.21,
                    "end": 1089.33,
                    "word": " of",
                    "probability": 0.90380859375
                },
                {
                    "start": 1089.33,
                    "end": 1089.65,
                    "word": " privacy",
                    "probability": 0.82763671875
                },
                {
                    "start": 1089.65,
                    "end": 1090.73,
                    "word": " implications",
                    "probability": 0.78369140625
                },
                {
                    "start": 1090.73,
                    "end": 1091.29,
                    "word": " come",
                    "probability": 0.783203125
                },
                {
                    "start": 1091.29,
                    "end": 1091.43,
                    "word": " out",
                    "probability": 0.82080078125
                },
                {
                    "start": 1091.43,
                    "end": 1091.55,
                    "word": " of",
                    "probability": 0.9013671875
                },
                {
                    "start": 1091.55,
                    "end": 1091.75,
                    "word": " that",
                    "probability": 0.884765625
                },
                {
                    "start": 1091.75,
                    "end": 1092.35,
                    "word": ".",
                    "probability": 0.9462890625
                }
            ]
        },
        {
            "id": 160,
            "seek": 109867,
            "start": 1093.13,
            "end": 1100.21,
            "text": " So the basis of differential privacy is really about quantifying the effect that a single user can have on output.",
            "tokens": [
                51464,
                407,
                264,
                5143,
                295,
                15756,
                11427,
                307,
                534,
                466,
                4426,
                5489,
                264,
                1802,
                300,
                257,
                2167,
                4195,
                393,
                362,
                322,
                5598,
                13,
                51814
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20032428697032748,
            "compression_ratio": 1.6738351254480286,
            "no_speech_prob": 0.0009255409240722656,
            "words": [
                {
                    "start": 1093.13,
                    "end": 1093.35,
                    "word": " So",
                    "probability": 0.65478515625
                },
                {
                    "start": 1093.35,
                    "end": 1095.07,
                    "word": " the",
                    "probability": 0.45068359375
                },
                {
                    "start": 1095.07,
                    "end": 1095.41,
                    "word": " basis",
                    "probability": 0.78466796875
                },
                {
                    "start": 1095.41,
                    "end": 1095.63,
                    "word": " of",
                    "probability": 0.90380859375
                },
                {
                    "start": 1095.63,
                    "end": 1095.93,
                    "word": " differential",
                    "probability": 0.74609375
                },
                {
                    "start": 1095.93,
                    "end": 1096.25,
                    "word": " privacy",
                    "probability": 0.83349609375
                },
                {
                    "start": 1096.25,
                    "end": 1096.49,
                    "word": " is",
                    "probability": 0.90234375
                },
                {
                    "start": 1096.49,
                    "end": 1096.75,
                    "word": " really",
                    "probability": 0.7744140625
                },
                {
                    "start": 1096.75,
                    "end": 1096.99,
                    "word": " about",
                    "probability": 0.80859375
                },
                {
                    "start": 1096.99,
                    "end": 1097.59,
                    "word": " quantifying",
                    "probability": 0.921142578125
                },
                {
                    "start": 1097.59,
                    "end": 1097.77,
                    "word": " the",
                    "probability": 0.8037109375
                },
                {
                    "start": 1097.77,
                    "end": 1098.07,
                    "word": " effect",
                    "probability": 0.78466796875
                },
                {
                    "start": 1098.07,
                    "end": 1098.23,
                    "word": " that",
                    "probability": 0.87646484375
                },
                {
                    "start": 1098.37,
                    "end": 1098.59,
                    "word": " a single",
                    "probability": 0.89111328125
                },
                {
                    "start": 1098.59,
                    "end": 1098.87,
                    "word": " user",
                    "probability": 0.79931640625
                },
                {
                    "start": 1098.87,
                    "end": 1099.09,
                    "word": " can",
                    "probability": 0.89794921875
                },
                {
                    "start": 1099.09,
                    "end": 1099.29,
                    "word": " have",
                    "probability": 0.876953125
                },
                {
                    "start": 1099.29,
                    "end": 1099.43,
                    "word": " on",
                    "probability": 0.90087890625
                },
                {
                    "start": 1099.43,
                    "end": 1099.75,
                    "word": " output",
                    "probability": 0.8388671875
                },
                {
                    "start": 1099.75,
                    "end": 1100.21,
                    "word": ".",
                    "probability": 0.6767578125
                }
            ]
        },
        {
            "id": 161,
            "seek": 112349,
            "start": 1100.71,
            "end": 1104.59,
            "text": " And part of a differentially private SQL engine really has to have the same property.",
            "tokens": [
                50364,
                400,
                644,
                295,
                257,
                819,
                2270,
                4551,
                19200,
                2848,
                534,
                575,
                281,
                362,
                264,
                912,
                4707,
                13,
                50514
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19884868170085707,
            "compression_ratio": 1.728395061728395,
            "no_speech_prob": 0.00572967529296875,
            "words": [
                {
                    "start": 1100.71,
                    "end": 1100.85,
                    "word": " And",
                    "probability": 0.052398681640625
                },
                {
                    "start": 1100.85,
                    "end": 1100.85,
                    "word": " part",
                    "probability": 0.178955078125
                },
                {
                    "start": 1100.85,
                    "end": 1100.97,
                    "word": " of",
                    "probability": 0.90576171875
                },
                {
                    "start": 1101.11,
                    "end": 1101.53,
                    "word": " a differentially",
                    "probability": 0.810546875
                },
                {
                    "start": 1101.53,
                    "end": 1101.83,
                    "word": " private",
                    "probability": 0.7744140625
                },
                {
                    "start": 1101.83,
                    "end": 1102.13,
                    "word": " SQL",
                    "probability": 0.849609375
                },
                {
                    "start": 1102.13,
                    "end": 1102.43,
                    "word": " engine",
                    "probability": 0.74755859375
                },
                {
                    "start": 1102.43,
                    "end": 1102.71,
                    "word": " really",
                    "probability": 0.7451171875
                },
                {
                    "start": 1102.71,
                    "end": 1102.95,
                    "word": " has",
                    "probability": 0.8994140625
                },
                {
                    "start": 1102.95,
                    "end": 1103.07,
                    "word": " to",
                    "probability": 0.90380859375
                },
                {
                    "start": 1103.07,
                    "end": 1103.23,
                    "word": " have",
                    "probability": 0.88916015625
                },
                {
                    "start": 1103.23,
                    "end": 1103.39,
                    "word": " the",
                    "probability": 0.8134765625
                },
                {
                    "start": 1103.39,
                    "end": 1103.63,
                    "word": " same",
                    "probability": 0.8037109375
                },
                {
                    "start": 1103.63,
                    "end": 1104.19,
                    "word": " property",
                    "probability": 0.7978515625
                },
                {
                    "start": 1104.19,
                    "end": 1104.59,
                    "word": ".",
                    "probability": 0.9833984375
                }
            ]
        },
        {
            "id": 162,
            "seek": 112349,
            "start": 1104.59,
            "end": 1111.07,
            "text": " So what we do is we actually split SQL statements into per-user transforms and cross-user aggregations.",
            "tokens": [
                50564,
                407,
                437,
                321,
                360,
                307,
                321,
                767,
                7472,
                19200,
                12363,
                666,
                680,
                12,
                18088,
                35592,
                293,
                3278,
                12,
                18088,
                16743,
                763,
                13,
                50864
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19884868170085707,
            "compression_ratio": 1.728395061728395,
            "no_speech_prob": 0.00572967529296875,
            "words": [
                {
                    "start": 1104.59,
                    "end": 1104.67,
                    "word": " So",
                    "probability": 0.87744140625
                },
                {
                    "start": 1104.67,
                    "end": 1104.81,
                    "word": " what",
                    "probability": 0.78125
                },
                {
                    "start": 1104.81,
                    "end": 1104.93,
                    "word": " we",
                    "probability": 0.9052734375
                },
                {
                    "start": 1104.93,
                    "end": 1105.09,
                    "word": " do",
                    "probability": 0.8955078125
                },
                {
                    "start": 1105.09,
                    "end": 1105.23,
                    "word": " is",
                    "probability": 0.88330078125
                },
                {
                    "start": 1105.23,
                    "end": 1105.37,
                    "word": " we",
                    "probability": 0.8740234375
                },
                {
                    "start": 1105.37,
                    "end": 1105.67,
                    "word": " actually",
                    "probability": 0.79736328125
                },
                {
                    "start": 1105.67,
                    "end": 1106.17,
                    "word": " split",
                    "probability": 0.85595703125
                },
                {
                    "start": 1106.17,
                    "end": 1106.47,
                    "word": " SQL",
                    "probability": 0.92919921875
                },
                {
                    "start": 1106.47,
                    "end": 1106.95,
                    "word": " statements",
                    "probability": 0.72265625
                },
                {
                    "start": 1106.95,
                    "end": 1107.45,
                    "word": " into",
                    "probability": 0.72314453125
                },
                {
                    "start": 1107.45,
                    "end": 1107.77,
                    "word": " per",
                    "probability": 0.9296875
                },
                {
                    "start": 1107.77,
                    "end": 1107.99,
                    "word": "-user",
                    "probability": 0.6712646484375
                },
                {
                    "start": 1107.99,
                    "end": 1108.45,
                    "word": " transforms",
                    "probability": 0.80517578125
                },
                {
                    "start": 1108.45,
                    "end": 1109.33,
                    "word": " and",
                    "probability": 0.83837890625
                },
                {
                    "start": 1109.33,
                    "end": 1109.71,
                    "word": " cross",
                    "probability": 0.8310546875
                },
                {
                    "start": 1109.71,
                    "end": 1109.95,
                    "word": "-user",
                    "probability": 0.900146484375
                },
                {
                    "start": 1109.95,
                    "end": 1110.65,
                    "word": " aggregations",
                    "probability": 0.884033203125
                },
                {
                    "start": 1110.65,
                    "end": 1111.07,
                    "word": ".",
                    "probability": 0.990234375
                }
            ]
        },
        {
            "id": 163,
            "seek": 112349,
            "start": 1111.07,
            "end": 1116.75,
            "text": " The per-user transforms are very generous, you can do essentially arbitrary, even Turing-complete SQL.",
            "tokens": [
                50914,
                440,
                680,
                12,
                18088,
                35592,
                366,
                588,
                14537,
                11,
                291,
                393,
                360,
                4476,
                23211,
                11,
                754,
                314,
                1345,
                12,
                1112,
                17220,
                19200,
                13,
                51114
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19884868170085707,
            "compression_ratio": 1.728395061728395,
            "no_speech_prob": 0.00572967529296875,
            "words": [
                {
                    "start": 1111.07,
                    "end": 1111.43,
                    "word": " The",
                    "probability": 0.8232421875
                },
                {
                    "start": 1111.43,
                    "end": 1111.61,
                    "word": " per",
                    "probability": 0.90283203125
                },
                {
                    "start": 1111.61,
                    "end": 1111.79,
                    "word": "-user",
                    "probability": 0.93994140625
                },
                {
                    "start": 1111.79,
                    "end": 1112.21,
                    "word": " transforms",
                    "probability": 0.7978515625
                },
                {
                    "start": 1112.21,
                    "end": 1112.49,
                    "word": " are",
                    "probability": 0.890625
                },
                {
                    "start": 1112.49,
                    "end": 1112.67,
                    "word": " very",
                    "probability": 0.74462890625
                },
                {
                    "start": 1112.67,
                    "end": 1113.05,
                    "word": " generous",
                    "probability": 0.84765625
                },
                {
                    "start": 1113.05,
                    "end": 1113.21,
                    "word": ",",
                    "probability": 0.244384765625
                },
                {
                    "start": 1113.21,
                    "end": 1113.31,
                    "word": " you",
                    "probability": 0.88134765625
                },
                {
                    "start": 1113.31,
                    "end": 1113.47,
                    "word": " can",
                    "probability": 0.90234375
                },
                {
                    "start": 1113.47,
                    "end": 1113.67,
                    "word": " do",
                    "probability": 0.9052734375
                },
                {
                    "start": 1113.67,
                    "end": 1114.11,
                    "word": " essentially",
                    "probability": 0.63916015625
                },
                {
                    "start": 1114.11,
                    "end": 1114.77,
                    "word": " arbitrary",
                    "probability": 0.8095703125
                },
                {
                    "start": 1114.77,
                    "end": 1115.07,
                    "word": ",",
                    "probability": 0.65625
                },
                {
                    "start": 1115.07,
                    "end": 1115.31,
                    "word": " even",
                    "probability": 0.77587890625
                },
                {
                    "start": 1115.31,
                    "end": 1115.55,
                    "word": " Turing",
                    "probability": 0.79248046875
                },
                {
                    "start": 1115.55,
                    "end": 1115.87,
                    "word": "-complete",
                    "probability": 0.7455240885416666
                },
                {
                    "start": 1115.87,
                    "end": 1116.25,
                    "word": " SQL",
                    "probability": 0.876953125
                },
                {
                    "start": 1116.25,
                    "end": 1116.75,
                    "word": ".",
                    "probability": 0.92919921875
                }
            ]
        },
        {
            "id": 164,
            "seek": 112349,
            "start": 1116.75,
            "end": 1125.53,
            "text": " The cross-user transforms are a restricted set of aggregation functions that are much more protected from a privacy standpoint.",
            "tokens": [
                51164,
                440,
                3278,
                12,
                18088,
                35592,
                366,
                257,
                20608,
                992,
                295,
                16743,
                399,
                6828,
                300,
                366,
                709,
                544,
                10594,
                490,
                257,
                11427,
                15827,
                13,
                51564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19884868170085707,
            "compression_ratio": 1.728395061728395,
            "no_speech_prob": 0.00572967529296875,
            "words": [
                {
                    "start": 1116.75,
                    "end": 1116.99,
                    "word": " The",
                    "probability": 0.8115234375
                },
                {
                    "start": 1116.99,
                    "end": 1117.31,
                    "word": " cross",
                    "probability": 0.849609375
                },
                {
                    "start": 1117.31,
                    "end": 1117.51,
                    "word": "-user",
                    "probability": 0.930908203125
                },
                {
                    "start": 1117.51,
                    "end": 1118.09,
                    "word": " transforms",
                    "probability": 0.8037109375
                },
                {
                    "start": 1118.09,
                    "end": 1118.41,
                    "word": " are",
                    "probability": 0.849609375
                },
                {
                    "start": 1118.65,
                    "end": 1119.07,
                    "word": " a restricted",
                    "probability": 0.63720703125
                },
                {
                    "start": 1119.07,
                    "end": 1119.61,
                    "word": " set",
                    "probability": 0.89794921875
                },
                {
                    "start": 1119.61,
                    "end": 1120.21,
                    "word": " of",
                    "probability": 0.904296875
                },
                {
                    "start": 1120.21,
                    "end": 1121.25,
                    "word": " aggregation",
                    "probability": 0.881103515625
                },
                {
                    "start": 1121.25,
                    "end": 1121.71,
                    "word": " functions",
                    "probability": 0.86181640625
                },
                {
                    "start": 1121.71,
                    "end": 1122.27,
                    "word": " that",
                    "probability": 0.87353515625
                },
                {
                    "start": 1122.27,
                    "end": 1122.65,
                    "word": " are",
                    "probability": 0.8955078125
                },
                {
                    "start": 1122.65,
                    "end": 1123.03,
                    "word": " much",
                    "probability": 0.810546875
                },
                {
                    "start": 1123.03,
                    "end": 1123.23,
                    "word": " more",
                    "probability": 0.89111328125
                },
                {
                    "start": 1123.23,
                    "end": 1123.63,
                    "word": " protected",
                    "probability": 0.66357421875
                },
                {
                    "start": 1123.63,
                    "end": 1123.99,
                    "word": " from",
                    "probability": 0.8125
                },
                {
                    "start": 1124.19,
                    "end": 1124.49,
                    "word": " a privacy",
                    "probability": 0.8271484375
                },
                {
                    "start": 1124.49,
                    "end": 1124.93,
                    "word": " standpoint",
                    "probability": 0.75390625
                },
                {
                    "start": 1124.93,
                    "end": 1125.53,
                    "word": ".",
                    "probability": 0.99951171875
                }
            ]
        },
        {
            "id": 165,
            "seek": 114973,
            "start": 1132.1,
            "end": 1145.24,
            "text": " Finally, you really want to ensure correctness when you're talking about any type of privacy-sensitive infrastructure.",
            "tokens": [
                50364,
                6288,
                11,
                291,
                534,
                528,
                281,
                5586,
                3006,
                1287,
                562,
                291,
                434,
                1417,
                466,
                604,
                2010,
                295,
                11427,
                12,
                82,
                34465,
                6896,
                13,
                51214
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19228830813400208,
            "compression_ratio": 1.4804469273743017,
            "no_speech_prob": 0.0330810546875,
            "words": [
                {
                    "start": 1132.1,
                    "end": 1132.82,
                    "word": " Finally",
                    "probability": 0.35205078125
                },
                {
                    "start": 1132.82,
                    "end": 1133.4,
                    "word": ",",
                    "probability": 0.880859375
                },
                {
                    "start": 1133.4,
                    "end": 1135.3,
                    "word": " you",
                    "probability": 0.8779296875
                },
                {
                    "start": 1135.3,
                    "end": 1135.6,
                    "word": " really",
                    "probability": 0.7705078125
                },
                {
                    "start": 1135.6,
                    "end": 1135.94,
                    "word": " want",
                    "probability": 0.8271484375
                },
                {
                    "start": 1135.94,
                    "end": 1136.14,
                    "word": " to",
                    "probability": 0.90673828125
                },
                {
                    "start": 1136.14,
                    "end": 1136.7,
                    "word": " ensure",
                    "probability": 0.80029296875
                },
                {
                    "start": 1136.7,
                    "end": 1137.78,
                    "word": " correctness",
                    "probability": 0.93408203125
                },
                {
                    "start": 1137.78,
                    "end": 1138.36,
                    "word": " when",
                    "probability": 0.8525390625
                },
                {
                    "start": 1138.36,
                    "end": 1138.62,
                    "word": " you're",
                    "probability": 0.916259765625
                },
                {
                    "start": 1138.62,
                    "end": 1138.96,
                    "word": " talking",
                    "probability": 0.7666015625
                },
                {
                    "start": 1138.96,
                    "end": 1139.74,
                    "word": " about",
                    "probability": 0.81298828125
                },
                {
                    "start": 1139.74,
                    "end": 1141.44,
                    "word": " any",
                    "probability": 0.82177734375
                },
                {
                    "start": 1141.44,
                    "end": 1141.94,
                    "word": " type",
                    "probability": 0.92822265625
                },
                {
                    "start": 1141.94,
                    "end": 1142.22,
                    "word": " of",
                    "probability": 0.90576171875
                },
                {
                    "start": 1142.22,
                    "end": 1142.56,
                    "word": " privacy",
                    "probability": 0.8369140625
                },
                {
                    "start": 1142.56,
                    "end": 1143.76,
                    "word": "-sensitive",
                    "probability": 0.702392578125
                },
                {
                    "start": 1143.76,
                    "end": 1144.52,
                    "word": " infrastructure",
                    "probability": 0.83056640625
                },
                {
                    "start": 1144.52,
                    "end": 1145.24,
                    "word": ".",
                    "probability": 0.76513671875
                }
            ]
        },
        {
            "id": 166,
            "seek": 114973,
            "start": 1145.54,
            "end": 1151.4,
            "text": " And unfortunately, proving that something you've built is differentially private is really hard.",
            "tokens": [
                51264,
                400,
                7015,
                11,
                27221,
                300,
                746,
                291,
                600,
                3094,
                307,
                819,
                2270,
                4551,
                307,
                534,
                1152,
                13,
                51514
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19228830813400208,
            "compression_ratio": 1.4804469273743017,
            "no_speech_prob": 0.0330810546875,
            "words": [
                {
                    "start": 1145.54,
                    "end": 1145.98,
                    "word": " And",
                    "probability": 0.73828125
                },
                {
                    "start": 1145.98,
                    "end": 1147.38,
                    "word": " unfortunately",
                    "probability": 0.72119140625
                },
                {
                    "start": 1147.38,
                    "end": 1147.56,
                    "word": ",",
                    "probability": 0.82568359375
                },
                {
                    "start": 1147.56,
                    "end": 1147.94,
                    "word": " proving",
                    "probability": 0.77294921875
                },
                {
                    "start": 1147.94,
                    "end": 1148.74,
                    "word": " that",
                    "probability": 0.87939453125
                },
                {
                    "start": 1148.74,
                    "end": 1149.1,
                    "word": " something",
                    "probability": 0.73974609375
                },
                {
                    "start": 1149.1,
                    "end": 1149.4,
                    "word": " you've",
                    "probability": 0.88671875
                },
                {
                    "start": 1149.4,
                    "end": 1149.58,
                    "word": " built",
                    "probability": 0.921875
                },
                {
                    "start": 1149.58,
                    "end": 1149.72,
                    "word": " is",
                    "probability": 0.89453125
                },
                {
                    "start": 1149.72,
                    "end": 1150.1,
                    "word": " differentially",
                    "probability": 0.88916015625
                },
                {
                    "start": 1150.1,
                    "end": 1150.4,
                    "word": " private",
                    "probability": 0.78662109375
                },
                {
                    "start": 1150.4,
                    "end": 1150.68,
                    "word": " is",
                    "probability": 0.892578125
                },
                {
                    "start": 1150.68,
                    "end": 1150.9,
                    "word": " really",
                    "probability": 0.78759765625
                },
                {
                    "start": 1150.9,
                    "end": 1151.3,
                    "word": " hard",
                    "probability": 0.8212890625
                },
                {
                    "start": 1151.3,
                    "end": 1151.4,
                    "word": ".",
                    "probability": 0.9599609375
                }
            ]
        },
        {
            "id": 167,
            "seek": 114973,
            "start": 1151.4,
            "end": 1154.1,
            "text": " It's so hard, it's actually provably undecidable.",
            "tokens": [
                51564,
                467,
                311,
                370,
                1152,
                11,
                309,
                311,
                767,
                1439,
                1188,
                674,
                3045,
                38089,
                13,
                51664
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19228830813400208,
            "compression_ratio": 1.4804469273743017,
            "no_speech_prob": 0.0330810546875,
            "words": [
                {
                    "start": 1151.4,
                    "end": 1151.68,
                    "word": " It's",
                    "probability": 0.951416015625
                },
                {
                    "start": 1151.68,
                    "end": 1151.88,
                    "word": " so",
                    "probability": 0.9052734375
                },
                {
                    "start": 1151.88,
                    "end": 1152.12,
                    "word": " hard",
                    "probability": 0.82275390625
                },
                {
                    "start": 1152.12,
                    "end": 1152.18,
                    "word": ",",
                    "probability": 0.60009765625
                },
                {
                    "start": 1152.18,
                    "end": 1152.32,
                    "word": " it's",
                    "probability": 0.950927734375
                },
                {
                    "start": 1152.32,
                    "end": 1152.56,
                    "word": " actually",
                    "probability": 0.7880859375
                },
                {
                    "start": 1152.56,
                    "end": 1153.08,
                    "word": " provably",
                    "probability": 0.963134765625
                },
                {
                    "start": 1153.08,
                    "end": 1153.74,
                    "word": " undecidable",
                    "probability": 0.8688151041666666
                },
                {
                    "start": 1153.74,
                    "end": 1154.1,
                    "word": ".",
                    "probability": 0.99560546875
                }
            ]
        },
        {
            "id": 168,
            "seek": 117973,
            "start": 1154.82,
            "end": 1162.84,
            "text": " That said, as part of the foundational libraries, we employ stochastic model checking as part of regression testing.",
            "tokens": [
                50364,
                663,
                848,
                11,
                382,
                644,
                295,
                264,
                32195,
                15148,
                11,
                321,
                3188,
                342,
                8997,
                2750,
                2316,
                8568,
                382,
                644,
                295,
                24590,
                4997,
                13,
                50764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.22148840697770267,
            "compression_ratio": 1.6619718309859155,
            "no_speech_prob": 0.06951904296875,
            "words": [
                {
                    "start": 1154.82,
                    "end": 1155.5,
                    "word": " That",
                    "probability": 0.7001953125
                },
                {
                    "start": 1155.5,
                    "end": 1155.92,
                    "word": " said",
                    "probability": 0.87255859375
                },
                {
                    "start": 1155.92,
                    "end": 1156.14,
                    "word": ",",
                    "probability": 0.88671875
                },
                {
                    "start": 1156.14,
                    "end": 1156.3,
                    "word": " as",
                    "probability": 0.90087890625
                },
                {
                    "start": 1156.3,
                    "end": 1156.48,
                    "word": " part",
                    "probability": 0.814453125
                },
                {
                    "start": 1156.48,
                    "end": 1156.54,
                    "word": " of",
                    "probability": 0.904296875
                },
                {
                    "start": 1156.54,
                    "end": 1156.64,
                    "word": " the",
                    "probability": 0.81640625
                },
                {
                    "start": 1156.64,
                    "end": 1157.08,
                    "word": " foundational",
                    "probability": 0.8525390625
                },
                {
                    "start": 1157.08,
                    "end": 1157.8,
                    "word": " libraries",
                    "probability": 0.892578125
                },
                {
                    "start": 1157.8,
                    "end": 1158.0,
                    "word": ",",
                    "probability": 0.8076171875
                },
                {
                    "start": 1158.0,
                    "end": 1158.08,
                    "word": " we",
                    "probability": 0.9052734375
                },
                {
                    "start": 1158.08,
                    "end": 1158.5,
                    "word": " employ",
                    "probability": 0.64501953125
                },
                {
                    "start": 1158.5,
                    "end": 1159.16,
                    "word": " stochastic",
                    "probability": 0.9344075520833334
                },
                {
                    "start": 1159.16,
                    "end": 1159.5,
                    "word": " model",
                    "probability": 0.8056640625
                },
                {
                    "start": 1159.5,
                    "end": 1159.88,
                    "word": " checking",
                    "probability": 0.76123046875
                },
                {
                    "start": 1159.88,
                    "end": 1160.84,
                    "word": " as",
                    "probability": 0.87841796875
                },
                {
                    "start": 1160.84,
                    "end": 1161.34,
                    "word": " part",
                    "probability": 0.83154296875
                },
                {
                    "start": 1161.34,
                    "end": 1161.44,
                    "word": " of",
                    "probability": 0.90625
                },
                {
                    "start": 1161.44,
                    "end": 1161.78,
                    "word": " regression",
                    "probability": 0.87060546875
                },
                {
                    "start": 1161.78,
                    "end": 1162.44,
                    "word": " testing",
                    "probability": 0.7431640625
                },
                {
                    "start": 1162.44,
                    "end": 1162.84,
                    "word": ".",
                    "probability": 0.9873046875
                }
            ]
        },
        {
            "id": 169,
            "seek": 117973,
            "start": 1162.84,
            "end": 1167.46,
            "text": " Now obviously this isn't a sufficient guarantee for differential privacy by itself,",
            "tokens": [
                50814,
                823,
                2745,
                341,
                1943,
                380,
                257,
                11563,
                10815,
                337,
                15756,
                11427,
                538,
                2564,
                11,
                51014
            ],
            "temperature": 0.0,
            "avg_logprob": -0.22148840697770267,
            "compression_ratio": 1.6619718309859155,
            "no_speech_prob": 0.06951904296875,
            "words": [
                {
                    "start": 1162.84,
                    "end": 1162.94,
                    "word": " Now",
                    "probability": 0.869140625
                },
                {
                    "start": 1162.94,
                    "end": 1163.22,
                    "word": " obviously",
                    "probability": 0.322021484375
                },
                {
                    "start": 1163.22,
                    "end": 1163.62,
                    "word": " this",
                    "probability": 0.454833984375
                },
                {
                    "start": 1163.62,
                    "end": 1163.9,
                    "word": " isn't",
                    "probability": 0.920654296875
                },
                {
                    "start": 1164.08,
                    "end": 1164.4,
                    "word": " a sufficient",
                    "probability": 0.81396484375
                },
                {
                    "start": 1164.4,
                    "end": 1164.94,
                    "word": " guarantee",
                    "probability": 0.83740234375
                },
                {
                    "start": 1164.94,
                    "end": 1165.92,
                    "word": " for",
                    "probability": 0.8818359375
                },
                {
                    "start": 1165.92,
                    "end": 1166.26,
                    "word": " differential",
                    "probability": 0.69775390625
                },
                {
                    "start": 1166.26,
                    "end": 1166.74,
                    "word": " privacy",
                    "probability": 0.84130859375
                },
                {
                    "start": 1166.74,
                    "end": 1166.98,
                    "word": " by",
                    "probability": 0.9013671875
                },
                {
                    "start": 1166.98,
                    "end": 1167.46,
                    "word": " itself",
                    "probability": 0.751953125
                },
                {
                    "start": 1167.46,
                    "end": 1167.46,
                    "word": ",",
                    "probability": 0.80078125
                }
            ]
        },
        {
            "id": 170,
            "seek": 117973,
            "start": 1167.46,
            "end": 1177.9,
            "text": " but it really quickly catches violations, both obvious and subtle, and provides a stable backdrop to more traditional privacy verification work and expert analysis.",
            "tokens": [
                51064,
                457,
                309,
                534,
                2661,
                25496,
                30405,
                11,
                1293,
                6322,
                293,
                13743,
                11,
                293,
                6417,
                257,
                8351,
                32697,
                281,
                544,
                5164,
                11427,
                30206,
                589,
                293,
                5844,
                5215,
                13,
                51514
            ],
            "temperature": 0.0,
            "avg_logprob": -0.22148840697770267,
            "compression_ratio": 1.6619718309859155,
            "no_speech_prob": 0.06951904296875,
            "words": [
                {
                    "start": 1167.46,
                    "end": 1168.02,
                    "word": " but",
                    "probability": 0.89501953125
                },
                {
                    "start": 1168.02,
                    "end": 1168.14,
                    "word": " it",
                    "probability": 0.900390625
                },
                {
                    "start": 1168.14,
                    "end": 1168.34,
                    "word": " really",
                    "probability": 0.78857421875
                },
                {
                    "start": 1168.34,
                    "end": 1168.78,
                    "word": " quickly",
                    "probability": 0.8369140625
                },
                {
                    "start": 1168.78,
                    "end": 1169.2,
                    "word": " catches",
                    "probability": 0.79931640625
                },
                {
                    "start": 1169.2,
                    "end": 1169.84,
                    "word": " violations",
                    "probability": 0.86865234375
                },
                {
                    "start": 1169.84,
                    "end": 1170.06,
                    "word": ",",
                    "probability": 0.6513671875
                },
                {
                    "start": 1170.06,
                    "end": 1170.22,
                    "word": " both",
                    "probability": 0.8291015625
                },
                {
                    "start": 1170.22,
                    "end": 1170.6,
                    "word": " obvious",
                    "probability": 0.85693359375
                },
                {
                    "start": 1170.6,
                    "end": 1170.88,
                    "word": " and",
                    "probability": 0.8935546875
                },
                {
                    "start": 1170.88,
                    "end": 1171.16,
                    "word": " subtle",
                    "probability": 0.73974609375
                },
                {
                    "start": 1171.16,
                    "end": 1171.38,
                    "word": ",",
                    "probability": 0.7783203125
                },
                {
                    "start": 1171.38,
                    "end": 1171.46,
                    "word": " and",
                    "probability": 0.8916015625
                },
                {
                    "start": 1171.46,
                    "end": 1172.0,
                    "word": " provides",
                    "probability": 0.1380615234375
                },
                {
                    "start": 1172.44,
                    "end": 1172.74,
                    "word": " a stable",
                    "probability": 0.80419921875
                },
                {
                    "start": 1172.74,
                    "end": 1173.44,
                    "word": " backdrop",
                    "probability": 0.76953125
                },
                {
                    "start": 1173.44,
                    "end": 1173.84,
                    "word": " to",
                    "probability": 0.90625
                },
                {
                    "start": 1173.84,
                    "end": 1174.2,
                    "word": " more",
                    "probability": 0.7236328125
                },
                {
                    "start": 1174.2,
                    "end": 1174.76,
                    "word": " traditional",
                    "probability": 0.7783203125
                },
                {
                    "start": 1174.76,
                    "end": 1175.22,
                    "word": " privacy",
                    "probability": 0.810546875
                },
                {
                    "start": 1175.22,
                    "end": 1175.68,
                    "word": " verification",
                    "probability": 0.9208984375
                },
                {
                    "start": 1175.68,
                    "end": 1176.24,
                    "word": " work",
                    "probability": 0.80126953125
                },
                {
                    "start": 1176.24,
                    "end": 1176.42,
                    "word": " and",
                    "probability": 0.81494140625
                },
                {
                    "start": 1176.42,
                    "end": 1176.74,
                    "word": " expert",
                    "probability": 0.8037109375
                },
                {
                    "start": 1176.74,
                    "end": 1177.22,
                    "word": " analysis",
                    "probability": 0.75830078125
                },
                {
                    "start": 1177.22,
                    "end": 1177.9,
                    "word": ".",
                    "probability": 0.99169921875
                }
            ]
        },
        {
            "id": 171,
            "seek": 117973,
            "start": 1178.16,
            "end": 1184.08,
            "text": " So this is just an example of a differentially private query, we'll get into this a little bit more later.",
            "tokens": [
                51564,
                407,
                341,
                307,
                445,
                364,
                1365,
                295,
                257,
                819,
                2270,
                4551,
                14581,
                11,
                321,
                603,
                483,
                666,
                341,
                257,
                707,
                857,
                544,
                1780,
                13,
                51814
            ],
            "temperature": 0.0,
            "avg_logprob": -0.22148840697770267,
            "compression_ratio": 1.6619718309859155,
            "no_speech_prob": 0.06951904296875,
            "words": [
                {
                    "start": 1178.16,
                    "end": 1178.82,
                    "word": " So",
                    "probability": 0.5908203125
                },
                {
                    "start": 1178.82,
                    "end": 1179.64,
                    "word": " this",
                    "probability": 0.6787109375
                },
                {
                    "start": 1179.64,
                    "end": 1179.84,
                    "word": " is",
                    "probability": 0.90673828125
                },
                {
                    "start": 1179.84,
                    "end": 1180.24,
                    "word": " just",
                    "probability": 0.87109375
                },
                {
                    "start": 1180.24,
                    "end": 1180.4,
                    "word": " an",
                    "probability": 0.904296875
                },
                {
                    "start": 1180.4,
                    "end": 1180.76,
                    "word": " example",
                    "probability": 0.93505859375
                },
                {
                    "start": 1180.76,
                    "end": 1181.16,
                    "word": " of",
                    "probability": 0.90380859375
                },
                {
                    "start": 1181.28,
                    "end": 1181.62,
                    "word": " a differentially",
                    "probability": 0.8388671875
                },
                {
                    "start": 1181.62,
                    "end": 1181.88,
                    "word": " private",
                    "probability": 0.783203125
                },
                {
                    "start": 1181.88,
                    "end": 1182.22,
                    "word": " query",
                    "probability": 0.88525390625
                },
                {
                    "start": 1182.22,
                    "end": 1182.34,
                    "word": ",",
                    "probability": 0.249755859375
                },
                {
                    "start": 1182.34,
                    "end": 1182.44,
                    "word": " we'll",
                    "probability": 0.90771484375
                },
                {
                    "start": 1182.44,
                    "end": 1182.6,
                    "word": " get",
                    "probability": 0.8994140625
                },
                {
                    "start": 1182.6,
                    "end": 1182.74,
                    "word": " into",
                    "probability": 0.72705078125
                },
                {
                    "start": 1182.74,
                    "end": 1182.94,
                    "word": " this",
                    "probability": 0.87744140625
                },
                {
                    "start": 1183.1,
                    "end": 1183.22,
                    "word": " a little",
                    "probability": 0.8037109375
                },
                {
                    "start": 1183.22,
                    "end": 1183.38,
                    "word": " bit",
                    "probability": 0.87548828125
                },
                {
                    "start": 1183.38,
                    "end": 1183.56,
                    "word": " more",
                    "probability": 0.8837890625
                },
                {
                    "start": 1183.56,
                    "end": 1183.86,
                    "word": " later",
                    "probability": 0.875
                },
                {
                    "start": 1183.86,
                    "end": 1184.08,
                    "word": ".",
                    "probability": 0.66162109375
                }
            ]
        },
        {
            "id": 172,
            "seek": 120863,
            "start": 1184.1,
            "end": 1191.28,
            "text": " It's meant to be user-centric, it highlights the fact that you're doing an anonymous query right up front.",
            "tokens": [
                50364,
                467,
                311,
                4140,
                281,
                312,
                4195,
                12,
                45300,
                11,
                309,
                14254,
                264,
                1186,
                300,
                291,
                434,
                884,
                364,
                24932,
                14581,
                558,
                493,
                1868,
                13,
                50664
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2180146988700418,
            "compression_ratio": 1.5308641975308641,
            "no_speech_prob": 0.01342010498046875,
            "words": [
                {
                    "start": 1184.1,
                    "end": 1185.2,
                    "word": " It's",
                    "probability": 0.87646484375
                },
                {
                    "start": 1185.2,
                    "end": 1185.4,
                    "word": " meant",
                    "probability": 0.86474609375
                },
                {
                    "start": 1185.4,
                    "end": 1185.54,
                    "word": " to",
                    "probability": 0.9072265625
                },
                {
                    "start": 1185.54,
                    "end": 1185.68,
                    "word": " be",
                    "probability": 0.904296875
                },
                {
                    "start": 1185.68,
                    "end": 1185.94,
                    "word": " user",
                    "probability": 0.7841796875
                },
                {
                    "start": 1185.94,
                    "end": 1186.32,
                    "word": "-centric",
                    "probability": 0.68359375
                },
                {
                    "start": 1186.32,
                    "end": 1186.64,
                    "word": ",",
                    "probability": 0.377197265625
                },
                {
                    "start": 1186.64,
                    "end": 1186.78,
                    "word": " it",
                    "probability": 0.8935546875
                },
                {
                    "start": 1186.78,
                    "end": 1187.12,
                    "word": " highlights",
                    "probability": 0.7001953125
                },
                {
                    "start": 1187.12,
                    "end": 1187.32,
                    "word": " the",
                    "probability": 0.810546875
                },
                {
                    "start": 1187.32,
                    "end": 1187.48,
                    "word": " fact",
                    "probability": 0.8125
                },
                {
                    "start": 1187.48,
                    "end": 1187.64,
                    "word": " that",
                    "probability": 0.88427734375
                },
                {
                    "start": 1187.64,
                    "end": 1187.78,
                    "word": " you're",
                    "probability": 0.91796875
                },
                {
                    "start": 1187.78,
                    "end": 1188.04,
                    "word": " doing",
                    "probability": 0.91552734375
                },
                {
                    "start": 1188.04,
                    "end": 1188.18,
                    "word": " an",
                    "probability": 0.90966796875
                },
                {
                    "start": 1188.18,
                    "end": 1188.52,
                    "word": " anonymous",
                    "probability": 0.71337890625
                },
                {
                    "start": 1188.52,
                    "end": 1188.98,
                    "word": " query",
                    "probability": 0.9013671875
                },
                {
                    "start": 1188.98,
                    "end": 1190.08,
                    "word": " right",
                    "probability": 0.75341796875
                },
                {
                    "start": 1190.08,
                    "end": 1190.28,
                    "word": " up",
                    "probability": 0.84423828125
                },
                {
                    "start": 1190.28,
                    "end": 1190.64,
                    "word": " front",
                    "probability": 0.85107421875
                },
                {
                    "start": 1190.64,
                    "end": 1191.28,
                    "word": ".",
                    "probability": 0.5986328125
                }
            ]
        },
        {
            "id": 173,
            "seek": 120863,
            "start": 1191.3,
            "end": 1202.32,
            "text": " And in general, it basically looks like a normal SQL query with just a few more keywords and syntactic sugar.",
            "tokens": [
                50714,
                400,
                294,
                2674,
                11,
                309,
                1936,
                1542,
                411,
                257,
                2710,
                19200,
                14581,
                365,
                445,
                257,
                1326,
                544,
                21009,
                293,
                23980,
                19892,
                5076,
                13,
                51214
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2180146988700418,
            "compression_ratio": 1.5308641975308641,
            "no_speech_prob": 0.01342010498046875,
            "words": [
                {
                    "start": 1191.3,
                    "end": 1191.54,
                    "word": " And",
                    "probability": 0.83642578125
                },
                {
                    "start": 1191.54,
                    "end": 1191.74,
                    "word": " in",
                    "probability": 0.8505859375
                },
                {
                    "start": 1191.74,
                    "end": 1192.08,
                    "word": " general",
                    "probability": 0.80615234375
                },
                {
                    "start": 1192.08,
                    "end": 1192.22,
                    "word": ",",
                    "probability": 0.90380859375
                },
                {
                    "start": 1192.22,
                    "end": 1192.4,
                    "word": " it",
                    "probability": 0.90380859375
                },
                {
                    "start": 1192.4,
                    "end": 1193.08,
                    "word": " basically",
                    "probability": 0.8857421875
                },
                {
                    "start": 1193.08,
                    "end": 1193.62,
                    "word": " looks",
                    "probability": 0.79150390625
                },
                {
                    "start": 1193.62,
                    "end": 1193.86,
                    "word": " like",
                    "probability": 0.88525390625
                },
                {
                    "start": 1194.04,
                    "end": 1194.42,
                    "word": " a normal",
                    "probability": 0.82421875
                },
                {
                    "start": 1194.42,
                    "end": 1196.38,
                    "word": " SQL",
                    "probability": 0.6728515625
                },
                {
                    "start": 1196.38,
                    "end": 1196.92,
                    "word": " query",
                    "probability": 0.91552734375
                },
                {
                    "start": 1196.92,
                    "end": 1197.92,
                    "word": " with",
                    "probability": 0.583984375
                },
                {
                    "start": 1197.92,
                    "end": 1198.14,
                    "word": " just",
                    "probability": 0.8876953125
                },
                {
                    "start": 1198.32,
                    "end": 1198.46,
                    "word": " a few",
                    "probability": 0.810546875
                },
                {
                    "start": 1198.46,
                    "end": 1199.02,
                    "word": " more",
                    "probability": 0.8955078125
                },
                {
                    "start": 1199.02,
                    "end": 1200.26,
                    "word": " keywords",
                    "probability": 0.58935546875
                },
                {
                    "start": 1200.26,
                    "end": 1200.62,
                    "word": " and",
                    "probability": 0.8828125
                },
                {
                    "start": 1200.62,
                    "end": 1201.16,
                    "word": " syntactic",
                    "probability": 0.814208984375
                },
                {
                    "start": 1201.16,
                    "end": 1201.68,
                    "word": " sugar",
                    "probability": 0.7724609375
                },
                {
                    "start": 1201.68,
                    "end": 1202.32,
                    "word": ".",
                    "probability": 0.98095703125
                }
            ]
        },
        {
            "id": 174,
            "seek": 120863,
            "start": 1202.62,
            "end": 1213.0,
            "text": " You see that this query includes things like count and average, and is selecting from some table in exactly the same way that you would perform normal SQL.",
            "tokens": [
                51264,
                509,
                536,
                300,
                341,
                14581,
                5974,
                721,
                411,
                1207,
                293,
                4274,
                11,
                293,
                307,
                18182,
                490,
                512,
                3199,
                294,
                2293,
                264,
                912,
                636,
                300,
                291,
                576,
                2042,
                2710,
                19200,
                13,
                51764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2180146988700418,
            "compression_ratio": 1.5308641975308641,
            "no_speech_prob": 0.01342010498046875,
            "words": [
                {
                    "start": 1202.62,
                    "end": 1202.78,
                    "word": " You",
                    "probability": 0.89013671875
                },
                {
                    "start": 1202.78,
                    "end": 1203.04,
                    "word": " see",
                    "probability": 0.818359375
                },
                {
                    "start": 1203.04,
                    "end": 1203.42,
                    "word": " that",
                    "probability": 0.86865234375
                },
                {
                    "start": 1203.42,
                    "end": 1204.1,
                    "word": " this",
                    "probability": 0.71533203125
                },
                {
                    "start": 1204.1,
                    "end": 1204.36,
                    "word": " query",
                    "probability": 0.8896484375
                },
                {
                    "start": 1204.36,
                    "end": 1204.82,
                    "word": " includes",
                    "probability": 0.8232421875
                },
                {
                    "start": 1204.82,
                    "end": 1205.16,
                    "word": " things",
                    "probability": 0.74365234375
                },
                {
                    "start": 1205.16,
                    "end": 1205.38,
                    "word": " like",
                    "probability": 0.87939453125
                },
                {
                    "start": 1205.38,
                    "end": 1205.7,
                    "word": " count",
                    "probability": 0.6796875
                },
                {
                    "start": 1205.7,
                    "end": 1205.86,
                    "word": " and",
                    "probability": 0.82958984375
                },
                {
                    "start": 1205.86,
                    "end": 1206.38,
                    "word": " average",
                    "probability": 0.671875
                },
                {
                    "start": 1206.38,
                    "end": 1206.9,
                    "word": ",",
                    "probability": 0.533203125
                },
                {
                    "start": 1206.9,
                    "end": 1207.24,
                    "word": " and",
                    "probability": 0.8955078125
                },
                {
                    "start": 1207.24,
                    "end": 1207.36,
                    "word": " is",
                    "probability": 0.44140625
                },
                {
                    "start": 1207.36,
                    "end": 1207.68,
                    "word": " selecting",
                    "probability": 0.70458984375
                },
                {
                    "start": 1207.68,
                    "end": 1208.12,
                    "word": " from",
                    "probability": 0.82275390625
                },
                {
                    "start": 1208.12,
                    "end": 1208.76,
                    "word": " some",
                    "probability": 0.81591796875
                },
                {
                    "start": 1208.76,
                    "end": 1209.6,
                    "word": " table",
                    "probability": 0.8115234375
                },
                {
                    "start": 1209.6,
                    "end": 1209.94,
                    "word": " in",
                    "probability": 0.84033203125
                },
                {
                    "start": 1209.94,
                    "end": 1210.48,
                    "word": " exactly",
                    "probability": 0.77490234375
                },
                {
                    "start": 1210.48,
                    "end": 1210.72,
                    "word": " the",
                    "probability": 0.81640625
                },
                {
                    "start": 1210.72,
                    "end": 1210.88,
                    "word": " same",
                    "probability": 0.818359375
                },
                {
                    "start": 1210.88,
                    "end": 1211.04,
                    "word": " way",
                    "probability": 0.89599609375
                },
                {
                    "start": 1211.04,
                    "end": 1211.18,
                    "word": " that",
                    "probability": 0.8798828125
                },
                {
                    "start": 1211.18,
                    "end": 1211.3,
                    "word": " you",
                    "probability": 0.89697265625
                },
                {
                    "start": 1211.3,
                    "end": 1211.54,
                    "word": " would",
                    "probability": 0.87060546875
                },
                {
                    "start": 1211.54,
                    "end": 1212.16,
                    "word": " perform",
                    "probability": 0.75830078125
                },
                {
                    "start": 1212.16,
                    "end": 1212.46,
                    "word": " normal",
                    "probability": 0.6728515625
                },
                {
                    "start": 1212.46,
                    "end": 1212.74,
                    "word": " SQL",
                    "probability": 0.931640625
                },
                {
                    "start": 1212.74,
                    "end": 1213.0,
                    "word": ".",
                    "probability": 0.99853515625
                }
            ]
        },
        {
            "id": 175,
            "seek": 123325,
            "start": 1215.28,
            "end": 1220.02,
            "text": " So that was a really cursory overview of this particular implementation.",
            "tokens": [
                50414,
                407,
                300,
                390,
                257,
                534,
                13946,
                827,
                12492,
                295,
                341,
                1729,
                11420,
                13,
                50664
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2027529776096344,
            "compression_ratio": 1.5720524017467248,
            "no_speech_prob": 0.011505126953125,
            "words": [
                {
                    "start": 1215.28,
                    "end": 1215.72,
                    "word": " So",
                    "probability": 0.5419921875
                },
                {
                    "start": 1215.72,
                    "end": 1216.16,
                    "word": " that",
                    "probability": 0.533203125
                },
                {
                    "start": 1216.16,
                    "end": 1216.3,
                    "word": " was",
                    "probability": 0.89453125
                },
                {
                    "start": 1216.42,
                    "end": 1216.6,
                    "word": " a really",
                    "probability": 0.78125
                },
                {
                    "start": 1216.6,
                    "end": 1217.2,
                    "word": " cursory",
                    "probability": 0.903564453125
                },
                {
                    "start": 1217.2,
                    "end": 1217.82,
                    "word": " overview",
                    "probability": 0.8349609375
                },
                {
                    "start": 1217.82,
                    "end": 1218.3,
                    "word": " of",
                    "probability": 0.8984375
                },
                {
                    "start": 1218.3,
                    "end": 1218.64,
                    "word": " this",
                    "probability": 0.84814453125
                },
                {
                    "start": 1218.64,
                    "end": 1219.08,
                    "word": " particular",
                    "probability": 0.8095703125
                },
                {
                    "start": 1219.08,
                    "end": 1219.58,
                    "word": " implementation",
                    "probability": 0.69091796875
                },
                {
                    "start": 1219.58,
                    "end": 1220.02,
                    "word": ".",
                    "probability": 0.8427734375
                }
            ]
        },
        {
            "id": 176,
            "seek": 123325,
            "start": 1220.2,
            "end": 1226.4,
            "text": " You can check out the details of our SQL implementation in full in the paper that we published at PETS this year.",
            "tokens": [
                50714,
                509,
                393,
                1520,
                484,
                264,
                4365,
                295,
                527,
                19200,
                11420,
                294,
                1577,
                294,
                264,
                3035,
                300,
                321,
                6572,
                412,
                21968,
                50,
                341,
                1064,
                13,
                51014
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2027529776096344,
            "compression_ratio": 1.5720524017467248,
            "no_speech_prob": 0.011505126953125,
            "words": [
                {
                    "start": 1220.2,
                    "end": 1220.6,
                    "word": " You",
                    "probability": 0.88427734375
                },
                {
                    "start": 1220.6,
                    "end": 1220.74,
                    "word": " can",
                    "probability": 0.89501953125
                },
                {
                    "start": 1220.74,
                    "end": 1220.88,
                    "word": " check",
                    "probability": 0.83984375
                },
                {
                    "start": 1220.88,
                    "end": 1221.0,
                    "word": " out",
                    "probability": 0.822265625
                },
                {
                    "start": 1221.0,
                    "end": 1221.12,
                    "word": " the",
                    "probability": 0.8125
                },
                {
                    "start": 1221.12,
                    "end": 1221.5,
                    "word": " details",
                    "probability": 0.701171875
                },
                {
                    "start": 1221.5,
                    "end": 1221.66,
                    "word": " of",
                    "probability": 0.900390625
                },
                {
                    "start": 1221.66,
                    "end": 1221.8,
                    "word": " our",
                    "probability": 0.798828125
                },
                {
                    "start": 1221.8,
                    "end": 1221.98,
                    "word": " SQL",
                    "probability": 0.77197265625
                },
                {
                    "start": 1221.98,
                    "end": 1222.7,
                    "word": " implementation",
                    "probability": 0.67822265625
                },
                {
                    "start": 1222.7,
                    "end": 1223.44,
                    "word": " in",
                    "probability": 0.89111328125
                },
                {
                    "start": 1223.44,
                    "end": 1223.8,
                    "word": " full",
                    "probability": 0.798828125
                },
                {
                    "start": 1223.8,
                    "end": 1224.02,
                    "word": " in",
                    "probability": 0.84912109375
                },
                {
                    "start": 1224.02,
                    "end": 1224.14,
                    "word": " the",
                    "probability": 0.80908203125
                },
                {
                    "start": 1224.14,
                    "end": 1224.36,
                    "word": " paper",
                    "probability": 0.8037109375
                },
                {
                    "start": 1224.36,
                    "end": 1224.56,
                    "word": " that",
                    "probability": 0.8583984375
                },
                {
                    "start": 1224.56,
                    "end": 1224.68,
                    "word": " we",
                    "probability": 0.9091796875
                },
                {
                    "start": 1224.68,
                    "end": 1225.02,
                    "word": " published",
                    "probability": 0.76220703125
                },
                {
                    "start": 1225.02,
                    "end": 1225.26,
                    "word": " at",
                    "probability": 0.880859375
                },
                {
                    "start": 1225.26,
                    "end": 1225.48,
                    "word": " PETS",
                    "probability": 0.640869140625
                },
                {
                    "start": 1225.48,
                    "end": 1225.7,
                    "word": " this",
                    "probability": 0.87109375
                },
                {
                    "start": 1225.7,
                    "end": 1225.96,
                    "word": " year",
                    "probability": 0.88525390625
                },
                {
                    "start": 1225.96,
                    "end": 1226.4,
                    "word": ".",
                    "probability": 0.689453125
                }
            ]
        },
        {
            "id": 177,
            "seek": 123325,
            "start": 1228.28,
            "end": 1232.6,
            "text": " And we'll be getting into some of the more detailed math later in the talk.",
            "tokens": [
                51064,
                400,
                321,
                603,
                312,
                1242,
                666,
                512,
                295,
                264,
                544,
                9942,
                5221,
                1780,
                294,
                264,
                751,
                13,
                51314
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2027529776096344,
            "compression_ratio": 1.5720524017467248,
            "no_speech_prob": 0.011505126953125,
            "words": [
                {
                    "start": 1228.28,
                    "end": 1228.72,
                    "word": " And",
                    "probability": 0.8251953125
                },
                {
                    "start": 1228.72,
                    "end": 1228.86,
                    "word": " we'll",
                    "probability": 0.922119140625
                },
                {
                    "start": 1228.86,
                    "end": 1229.02,
                    "word": " be",
                    "probability": 0.9033203125
                },
                {
                    "start": 1229.02,
                    "end": 1229.24,
                    "word": " getting",
                    "probability": 0.89697265625
                },
                {
                    "start": 1229.24,
                    "end": 1229.64,
                    "word": " into",
                    "probability": 0.7314453125
                },
                {
                    "start": 1229.64,
                    "end": 1230.46,
                    "word": " some",
                    "probability": 0.81396484375
                },
                {
                    "start": 1230.46,
                    "end": 1230.56,
                    "word": " of",
                    "probability": 0.8974609375
                },
                {
                    "start": 1230.56,
                    "end": 1230.64,
                    "word": " the",
                    "probability": 0.81396484375
                },
                {
                    "start": 1230.64,
                    "end": 1230.8,
                    "word": " more",
                    "probability": 0.8896484375
                },
                {
                    "start": 1230.8,
                    "end": 1231.12,
                    "word": " detailed",
                    "probability": 0.73681640625
                },
                {
                    "start": 1231.12,
                    "end": 1231.44,
                    "word": " math",
                    "probability": 0.90771484375
                },
                {
                    "start": 1231.44,
                    "end": 1231.66,
                    "word": " later",
                    "probability": 0.86865234375
                },
                {
                    "start": 1231.66,
                    "end": 1231.78,
                    "word": " in",
                    "probability": 0.90576171875
                },
                {
                    "start": 1231.78,
                    "end": 1231.9,
                    "word": " the",
                    "probability": 0.81396484375
                },
                {
                    "start": 1231.9,
                    "end": 1232.16,
                    "word": " talk",
                    "probability": 0.8056640625
                },
                {
                    "start": 1232.16,
                    "end": 1232.6,
                    "word": ".",
                    "probability": 0.9921875
                }
            ]
        },
        {
            "id": 178,
            "seek": 123325,
            "start": 1232.88,
            "end": 1237.62,
            "text": " So we're going to spend some time to return to the math behind differential privacy a little bit.",
            "tokens": [
                51364,
                407,
                321,
                434,
                516,
                281,
                3496,
                512,
                565,
                281,
                2736,
                281,
                264,
                5221,
                2261,
                15756,
                11427,
                257,
                707,
                857,
                13,
                51564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2027529776096344,
            "compression_ratio": 1.5720524017467248,
            "no_speech_prob": 0.011505126953125,
            "words": [
                {
                    "start": 1232.88,
                    "end": 1233.32,
                    "word": " So",
                    "probability": 0.87548828125
                },
                {
                    "start": 1233.32,
                    "end": 1233.46,
                    "word": " we're",
                    "probability": 0.91748046875
                },
                {
                    "start": 1233.46,
                    "end": 1233.54,
                    "word": " going",
                    "probability": 0.7470703125
                },
                {
                    "start": 1233.54,
                    "end": 1233.6,
                    "word": " to",
                    "probability": 0.9091796875
                },
                {
                    "start": 1233.6,
                    "end": 1233.78,
                    "word": " spend",
                    "probability": 0.81884765625
                },
                {
                    "start": 1233.78,
                    "end": 1234.02,
                    "word": " some",
                    "probability": 0.81982421875
                },
                {
                    "start": 1234.02,
                    "end": 1234.24,
                    "word": " time",
                    "probability": 0.81396484375
                },
                {
                    "start": 1234.24,
                    "end": 1234.4,
                    "word": " to",
                    "probability": 0.89990234375
                },
                {
                    "start": 1234.4,
                    "end": 1234.74,
                    "word": " return",
                    "probability": 0.79248046875
                },
                {
                    "start": 1234.74,
                    "end": 1235.06,
                    "word": " to",
                    "probability": 0.90673828125
                },
                {
                    "start": 1235.06,
                    "end": 1235.24,
                    "word": " the",
                    "probability": 0.81591796875
                },
                {
                    "start": 1235.24,
                    "end": 1235.44,
                    "word": " math",
                    "probability": 0.900390625
                },
                {
                    "start": 1235.44,
                    "end": 1235.8,
                    "word": " behind",
                    "probability": 0.853515625
                },
                {
                    "start": 1235.8,
                    "end": 1236.16,
                    "word": " differential",
                    "probability": 0.69580078125
                },
                {
                    "start": 1236.16,
                    "end": 1236.6,
                    "word": " privacy",
                    "probability": 0.82275390625
                },
                {
                    "start": 1236.8,
                    "end": 1236.9,
                    "word": " a little",
                    "probability": 0.80029296875
                },
                {
                    "start": 1236.9,
                    "end": 1237.18,
                    "word": " bit",
                    "probability": 0.8994140625
                },
                {
                    "start": 1237.18,
                    "end": 1237.62,
                    "word": ".",
                    "probability": 0.98583984375
                }
            ]
        },
        {
            "id": 179,
            "seek": 126061,
            "start": 1237.62,
            "end": 1243.52,
            "text": " We talked kind of about how our SQL engine was structured and a little bit about differential privacy.",
            "tokens": [
                50364,
                492,
                2825,
                733,
                295,
                466,
                577,
                527,
                19200,
                2848,
                390,
                18519,
                293,
                257,
                707,
                857,
                466,
                15756,
                11427,
                13,
                50614
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19039948453608246,
            "compression_ratio": 1.588235294117647,
            "no_speech_prob": 0.0838623046875,
            "words": [
                {
                    "start": 1237.62,
                    "end": 1238.2,
                    "word": " We",
                    "probability": 0.52001953125
                },
                {
                    "start": 1238.2,
                    "end": 1238.58,
                    "word": " talked",
                    "probability": 0.77294921875
                },
                {
                    "start": 1238.58,
                    "end": 1238.92,
                    "word": " kind",
                    "probability": 0.63232421875
                },
                {
                    "start": 1238.92,
                    "end": 1239.02,
                    "word": " of",
                    "probability": 0.9033203125
                },
                {
                    "start": 1239.02,
                    "end": 1239.42,
                    "word": " about",
                    "probability": 0.82275390625
                },
                {
                    "start": 1239.42,
                    "end": 1239.84,
                    "word": " how",
                    "probability": 0.8896484375
                },
                {
                    "start": 1239.84,
                    "end": 1240.0,
                    "word": " our",
                    "probability": 0.78515625
                },
                {
                    "start": 1240.0,
                    "end": 1240.22,
                    "word": " SQL",
                    "probability": 0.87646484375
                },
                {
                    "start": 1240.22,
                    "end": 1240.46,
                    "word": " engine",
                    "probability": 0.740234375
                },
                {
                    "start": 1240.46,
                    "end": 1240.78,
                    "word": " was",
                    "probability": 0.88232421875
                },
                {
                    "start": 1240.78,
                    "end": 1241.14,
                    "word": " structured",
                    "probability": 0.59423828125
                },
                {
                    "start": 1241.14,
                    "end": 1241.56,
                    "word": " and",
                    "probability": 0.78466796875
                },
                {
                    "start": 1241.7,
                    "end": 1241.82,
                    "word": " a little",
                    "probability": 0.802734375
                },
                {
                    "start": 1241.82,
                    "end": 1242.0,
                    "word": " bit",
                    "probability": 0.9033203125
                },
                {
                    "start": 1242.0,
                    "end": 1242.18,
                    "word": " about",
                    "probability": 0.822265625
                },
                {
                    "start": 1242.18,
                    "end": 1242.52,
                    "word": " differential",
                    "probability": 0.708984375
                },
                {
                    "start": 1242.52,
                    "end": 1242.96,
                    "word": " privacy",
                    "probability": 0.86279296875
                },
                {
                    "start": 1242.96,
                    "end": 1243.52,
                    "word": ".",
                    "probability": 0.57568359375
                }
            ]
        },
        {
            "id": 180,
            "seek": 126061,
            "start": 1243.56,
            "end": 1252.72,
            "text": " But going into a bit more of the details, you know, usually when a DP is introduced, it looks something like this.",
            "tokens": [
                50664,
                583,
                516,
                666,
                257,
                857,
                544,
                295,
                264,
                4365,
                11,
                291,
                458,
                11,
                2673,
                562,
                257,
                42796,
                307,
                7268,
                11,
                309,
                1542,
                746,
                411,
                341,
                13,
                51064
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19039948453608246,
            "compression_ratio": 1.588235294117647,
            "no_speech_prob": 0.0838623046875,
            "words": [
                {
                    "start": 1243.56,
                    "end": 1244.06,
                    "word": " But",
                    "probability": 0.861328125
                },
                {
                    "start": 1244.06,
                    "end": 1245.48,
                    "word": " going",
                    "probability": 0.74951171875
                },
                {
                    "start": 1245.48,
                    "end": 1245.82,
                    "word": " into",
                    "probability": 0.7333984375
                },
                {
                    "start": 1246.04,
                    "end": 1246.2,
                    "word": " a bit",
                    "probability": 0.89892578125
                },
                {
                    "start": 1246.2,
                    "end": 1246.48,
                    "word": " more",
                    "probability": 0.8828125
                },
                {
                    "start": 1246.48,
                    "end": 1246.62,
                    "word": " of",
                    "probability": 0.89892578125
                },
                {
                    "start": 1246.62,
                    "end": 1246.74,
                    "word": " the",
                    "probability": 0.8115234375
                },
                {
                    "start": 1246.74,
                    "end": 1247.28,
                    "word": " details",
                    "probability": 0.705078125
                },
                {
                    "start": 1247.28,
                    "end": 1248.4,
                    "word": ",",
                    "probability": 0.79443359375
                },
                {
                    "start": 1248.4,
                    "end": 1248.46,
                    "word": " you",
                    "probability": 0.4521484375
                },
                {
                    "start": 1248.46,
                    "end": 1248.56,
                    "word": " know",
                    "probability": 0.81494140625
                },
                {
                    "start": 1248.56,
                    "end": 1248.6,
                    "word": ",",
                    "probability": 0.91162109375
                },
                {
                    "start": 1248.6,
                    "end": 1248.88,
                    "word": " usually",
                    "probability": 0.8271484375
                },
                {
                    "start": 1248.88,
                    "end": 1249.12,
                    "word": " when",
                    "probability": 0.828125
                },
                {
                    "start": 1249.52,
                    "end": 1249.78,
                    "word": " a DP",
                    "probability": 0.71484375
                },
                {
                    "start": 1249.78,
                    "end": 1250.06,
                    "word": " is",
                    "probability": 0.9052734375
                },
                {
                    "start": 1250.06,
                    "end": 1250.6,
                    "word": " introduced",
                    "probability": 0.75439453125
                },
                {
                    "start": 1250.6,
                    "end": 1250.72,
                    "word": ",",
                    "probability": 0.89794921875
                },
                {
                    "start": 1250.72,
                    "end": 1250.94,
                    "word": " it",
                    "probability": 0.89892578125
                },
                {
                    "start": 1250.94,
                    "end": 1251.12,
                    "word": " looks",
                    "probability": 0.79296875
                },
                {
                    "start": 1251.12,
                    "end": 1251.42,
                    "word": " something",
                    "probability": 0.72265625
                },
                {
                    "start": 1251.42,
                    "end": 1251.7,
                    "word": " like",
                    "probability": 0.87646484375
                },
                {
                    "start": 1251.7,
                    "end": 1252.16,
                    "word": " this",
                    "probability": 0.8837890625
                },
                {
                    "start": 1252.16,
                    "end": 1252.72,
                    "word": ".",
                    "probability": 0.9599609375
                }
            ]
        },
        {
            "id": 181,
            "seek": 126061,
            "start": 1253.32,
            "end": 1259.1,
            "text": " Of course, what we're actually saying when we say some sort of probabilistic statement about any differentially private system",
            "tokens": [
                51114,
                2720,
                1164,
                11,
                437,
                321,
                434,
                767,
                1566,
                562,
                321,
                584,
                512,
                1333,
                295,
                31959,
                3142,
                5629,
                466,
                604,
                819,
                2270,
                4551,
                1185,
                51414
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19039948453608246,
            "compression_ratio": 1.588235294117647,
            "no_speech_prob": 0.0838623046875,
            "words": [
                {
                    "start": 1253.32,
                    "end": 1253.88,
                    "word": " Of",
                    "probability": 0.8505859375
                },
                {
                    "start": 1253.88,
                    "end": 1253.9,
                    "word": " course",
                    "probability": 0.85595703125
                },
                {
                    "start": 1253.9,
                    "end": 1254.04,
                    "word": ",",
                    "probability": 0.94580078125
                },
                {
                    "start": 1254.04,
                    "end": 1254.22,
                    "word": " what",
                    "probability": 0.84375
                },
                {
                    "start": 1254.22,
                    "end": 1254.34,
                    "word": " we're",
                    "probability": 0.92626953125
                },
                {
                    "start": 1254.34,
                    "end": 1254.58,
                    "word": " actually",
                    "probability": 0.76708984375
                },
                {
                    "start": 1254.58,
                    "end": 1255.02,
                    "word": " saying",
                    "probability": 0.8115234375
                },
                {
                    "start": 1255.02,
                    "end": 1255.32,
                    "word": " when",
                    "probability": 0.7177734375
                },
                {
                    "start": 1255.32,
                    "end": 1255.46,
                    "word": " we",
                    "probability": 0.90673828125
                },
                {
                    "start": 1255.46,
                    "end": 1255.62,
                    "word": " say",
                    "probability": 0.908203125
                },
                {
                    "start": 1255.62,
                    "end": 1255.78,
                    "word": " some",
                    "probability": 0.82666015625
                },
                {
                    "start": 1255.78,
                    "end": 1255.96,
                    "word": " sort",
                    "probability": 0.80615234375
                },
                {
                    "start": 1255.96,
                    "end": 1256.04,
                    "word": " of",
                    "probability": 0.90966796875
                },
                {
                    "start": 1256.04,
                    "end": 1256.66,
                    "word": " probabilistic",
                    "probability": 0.91259765625
                },
                {
                    "start": 1256.66,
                    "end": 1257.18,
                    "word": " statement",
                    "probability": 0.82373046875
                },
                {
                    "start": 1257.18,
                    "end": 1257.68,
                    "word": " about",
                    "probability": 0.81591796875
                },
                {
                    "start": 1257.68,
                    "end": 1257.98,
                    "word": " any",
                    "probability": 0.81591796875
                },
                {
                    "start": 1257.98,
                    "end": 1258.4,
                    "word": " differentially",
                    "probability": 0.788330078125
                },
                {
                    "start": 1258.4,
                    "end": 1258.64,
                    "word": " private",
                    "probability": 0.78466796875
                },
                {
                    "start": 1258.64,
                    "end": 1259.1,
                    "word": " system",
                    "probability": 0.865234375
                }
            ]
        },
        {
            "id": 182,
            "seek": 126061,
            "start": 1259.44,
            "end": 1264.98,
            "text": " is that adding or removing a user from a data set doesn't change the results that much.",
            "tokens": [
                51464,
                307,
                300,
                5127,
                420,
                12720,
                257,
                4195,
                490,
                257,
                1412,
                992,
                1177,
                380,
                1319,
                264,
                3542,
                300,
                709,
                13,
                51664
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19039948453608246,
            "compression_ratio": 1.588235294117647,
            "no_speech_prob": 0.0838623046875,
            "words": [
                {
                    "start": 1259.44,
                    "end": 1259.72,
                    "word": " is",
                    "probability": 0.6474609375
                },
                {
                    "start": 1259.72,
                    "end": 1259.92,
                    "word": " that",
                    "probability": 0.8818359375
                },
                {
                    "start": 1259.92,
                    "end": 1260.2,
                    "word": " adding",
                    "probability": 0.81787109375
                },
                {
                    "start": 1260.2,
                    "end": 1260.36,
                    "word": " or",
                    "probability": 0.84228515625
                },
                {
                    "start": 1260.36,
                    "end": 1260.74,
                    "word": " removing",
                    "probability": 0.7626953125
                },
                {
                    "start": 1261.1,
                    "end": 1261.34,
                    "word": " a user",
                    "probability": 0.81298828125
                },
                {
                    "start": 1261.34,
                    "end": 1261.54,
                    "word": " from",
                    "probability": 0.81494140625
                },
                {
                    "start": 1261.66,
                    "end": 1261.8,
                    "word": " a data",
                    "probability": 0.54296875
                },
                {
                    "start": 1261.8,
                    "end": 1262.06,
                    "word": " set",
                    "probability": 0.87158203125
                },
                {
                    "start": 1262.06,
                    "end": 1262.34,
                    "word": " doesn't",
                    "probability": 0.9296875
                },
                {
                    "start": 1262.34,
                    "end": 1262.7,
                    "word": " change",
                    "probability": 0.84912109375
                },
                {
                    "start": 1262.7,
                    "end": 1262.88,
                    "word": " the",
                    "probability": 0.80908203125
                },
                {
                    "start": 1262.88,
                    "end": 1263.32,
                    "word": " results",
                    "probability": 0.68798828125
                },
                {
                    "start": 1263.32,
                    "end": 1263.98,
                    "word": " that",
                    "probability": 0.8603515625
                },
                {
                    "start": 1263.98,
                    "end": 1264.42,
                    "word": " much",
                    "probability": 0.81689453125
                },
                {
                    "start": 1264.42,
                    "end": 1264.98,
                    "word": ".",
                    "probability": 0.96142578125
                }
            ]
        },
        {
            "id": 183,
            "seek": 128553,
            "start": 1264.98,
            "end": 1269.22,
            "text": " That much actually has a really rigorous technical definition.",
            "tokens": [
                50364,
                663,
                709,
                767,
                575,
                257,
                534,
                29882,
                6191,
                7123,
                13,
                50514
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18679775146955854,
            "compression_ratio": 1.5418502202643172,
            "no_speech_prob": 0.062744140625,
            "words": [
                {
                    "start": 1264.98,
                    "end": 1265.82,
                    "word": " That",
                    "probability": 0.352783203125
                },
                {
                    "start": 1265.82,
                    "end": 1266.06,
                    "word": " much",
                    "probability": 0.802734375
                },
                {
                    "start": 1266.06,
                    "end": 1266.32,
                    "word": " actually",
                    "probability": 0.7861328125
                },
                {
                    "start": 1266.32,
                    "end": 1266.48,
                    "word": " has",
                    "probability": 0.88916015625
                },
                {
                    "start": 1266.68,
                    "end": 1266.94,
                    "word": " a really",
                    "probability": 0.7734375
                },
                {
                    "start": 1266.94,
                    "end": 1267.52,
                    "word": " rigorous",
                    "probability": 0.861328125
                },
                {
                    "start": 1267.52,
                    "end": 1268.06,
                    "word": " technical",
                    "probability": 0.8203125
                },
                {
                    "start": 1268.06,
                    "end": 1268.66,
                    "word": " definition",
                    "probability": 0.84033203125
                },
                {
                    "start": 1268.66,
                    "end": 1269.22,
                    "word": ".",
                    "probability": 0.9462890625
                }
            ]
        },
        {
            "id": 184,
            "seek": 128553,
            "start": 1270.04,
            "end": 1272.8,
            "text": " There's two parameters, there's epsilon and delta.",
            "tokens": [
                50564,
                821,
                311,
                732,
                9834,
                11,
                456,
                311,
                17889,
                293,
                8289,
                13,
                50714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18679775146955854,
            "compression_ratio": 1.5418502202643172,
            "no_speech_prob": 0.062744140625,
            "words": [
                {
                    "start": 1270.04,
                    "end": 1270.6,
                    "word": " There's",
                    "probability": 0.841796875
                },
                {
                    "start": 1270.6,
                    "end": 1270.76,
                    "word": " two",
                    "probability": 0.8115234375
                },
                {
                    "start": 1270.76,
                    "end": 1271.12,
                    "word": " parameters",
                    "probability": 0.89453125
                },
                {
                    "start": 1271.12,
                    "end": 1271.38,
                    "word": ",",
                    "probability": 0.5107421875
                },
                {
                    "start": 1271.38,
                    "end": 1271.5,
                    "word": " there's",
                    "probability": 0.8994140625
                },
                {
                    "start": 1271.5,
                    "end": 1271.78,
                    "word": " epsilon",
                    "probability": 0.59130859375
                },
                {
                    "start": 1271.78,
                    "end": 1272.0,
                    "word": " and",
                    "probability": 0.880859375
                },
                {
                    "start": 1272.0,
                    "end": 1272.34,
                    "word": " delta",
                    "probability": 0.8046875
                },
                {
                    "start": 1272.34,
                    "end": 1272.8,
                    "word": ".",
                    "probability": 0.94384765625
                }
            ]
        },
        {
            "id": 185,
            "seek": 128553,
            "start": 1272.88,
            "end": 1278.38,
            "text": " Epsilon, we generally think about in a sort of a Bayesian context,",
            "tokens": [
                50764,
                462,
                16592,
                11,
                321,
                5101,
                519,
                466,
                294,
                257,
                1333,
                295,
                257,
                7840,
                42434,
                4319,
                11,
                50964
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18679775146955854,
            "compression_ratio": 1.5418502202643172,
            "no_speech_prob": 0.062744140625,
            "words": [
                {
                    "start": 1272.88,
                    "end": 1273.44,
                    "word": " Epsilon",
                    "probability": 0.884521484375
                },
                {
                    "start": 1273.44,
                    "end": 1274.02,
                    "word": ",",
                    "probability": 0.62451171875
                },
                {
                    "start": 1274.02,
                    "end": 1274.72,
                    "word": " we",
                    "probability": 0.88671875
                },
                {
                    "start": 1274.72,
                    "end": 1275.08,
                    "word": " generally",
                    "probability": 0.83447265625
                },
                {
                    "start": 1275.08,
                    "end": 1275.36,
                    "word": " think",
                    "probability": 0.80859375
                },
                {
                    "start": 1275.36,
                    "end": 1275.72,
                    "word": " about",
                    "probability": 0.81640625
                },
                {
                    "start": 1275.72,
                    "end": 1276.1,
                    "word": " in",
                    "probability": 0.87451171875
                },
                {
                    "start": 1276.42,
                    "end": 1276.58,
                    "word": " a sort",
                    "probability": 0.7939453125
                },
                {
                    "start": 1276.58,
                    "end": 1276.68,
                    "word": " of",
                    "probability": 0.90087890625
                },
                {
                    "start": 1276.8,
                    "end": 1277.1,
                    "word": " a Bayesian",
                    "probability": 0.9248046875
                },
                {
                    "start": 1277.1,
                    "end": 1277.86,
                    "word": " context",
                    "probability": 0.73486328125
                },
                {
                    "start": 1277.86,
                    "end": 1278.38,
                    "word": ",",
                    "probability": 0.6630859375
                }
            ]
        },
        {
            "id": 186,
            "seek": 128553,
            "start": 1278.38,
            "end": 1286.02,
            "text": " which is to say that supposing I have some notion of a detail of a user that you're trying to target, right?",
            "tokens": [
                51014,
                597,
                307,
                281,
                584,
                300,
                1003,
                6110,
                286,
                362,
                512,
                10710,
                295,
                257,
                2607,
                295,
                257,
                4195,
                300,
                291,
                434,
                1382,
                281,
                3779,
                11,
                558,
                30,
                51364
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18679775146955854,
            "compression_ratio": 1.5418502202643172,
            "no_speech_prob": 0.062744140625,
            "words": [
                {
                    "start": 1278.38,
                    "end": 1279.14,
                    "word": " which",
                    "probability": 0.91357421875
                },
                {
                    "start": 1279.14,
                    "end": 1279.28,
                    "word": " is",
                    "probability": 0.90478515625
                },
                {
                    "start": 1279.28,
                    "end": 1279.44,
                    "word": " to",
                    "probability": 0.90673828125
                },
                {
                    "start": 1279.44,
                    "end": 1279.68,
                    "word": " say",
                    "probability": 0.90234375
                },
                {
                    "start": 1279.68,
                    "end": 1280.22,
                    "word": " that",
                    "probability": 0.87451171875
                },
                {
                    "start": 1280.22,
                    "end": 1280.9,
                    "word": " supposing",
                    "probability": 0.863525390625
                },
                {
                    "start": 1280.9,
                    "end": 1281.18,
                    "word": " I",
                    "probability": 0.98876953125
                },
                {
                    "start": 1281.18,
                    "end": 1281.48,
                    "word": " have",
                    "probability": 0.88330078125
                },
                {
                    "start": 1281.48,
                    "end": 1281.76,
                    "word": " some",
                    "probability": 0.81787109375
                },
                {
                    "start": 1281.76,
                    "end": 1282.3,
                    "word": " notion",
                    "probability": 0.87939453125
                },
                {
                    "start": 1282.3,
                    "end": 1282.94,
                    "word": " of",
                    "probability": 0.8974609375
                },
                {
                    "start": 1283.08,
                    "end": 1283.4,
                    "word": " a detail",
                    "probability": 0.8330078125
                },
                {
                    "start": 1283.4,
                    "end": 1283.68,
                    "word": " of",
                    "probability": 0.900390625
                },
                {
                    "start": 1284.3,
                    "end": 1284.56,
                    "word": " a user",
                    "probability": 0.81005859375
                },
                {
                    "start": 1284.56,
                    "end": 1284.76,
                    "word": " that",
                    "probability": 0.884765625
                },
                {
                    "start": 1284.76,
                    "end": 1284.9,
                    "word": " you're",
                    "probability": 0.922607421875
                },
                {
                    "start": 1284.9,
                    "end": 1285.1,
                    "word": " trying",
                    "probability": 0.80126953125
                },
                {
                    "start": 1285.1,
                    "end": 1285.28,
                    "word": " to",
                    "probability": 0.9013671875
                },
                {
                    "start": 1285.28,
                    "end": 1285.58,
                    "word": " target",
                    "probability": 0.84765625
                },
                {
                    "start": 1285.58,
                    "end": 1285.78,
                    "word": ",",
                    "probability": 0.76513671875
                },
                {
                    "start": 1285.78,
                    "end": 1285.92,
                    "word": " right",
                    "probability": 0.8310546875
                },
                {
                    "start": 1285.92,
                    "end": 1286.02,
                    "word": "?",
                    "probability": 0.66064453125
                }
            ]
        },
        {
            "id": 187,
            "seek": 128553,
            "start": 1286.02,
            "end": 1289.9,
            "text": " Supposing you're about 60% sure that they live in Wisconsin,",
            "tokens": [
                51414,
                9391,
                6110,
                291,
                434,
                466,
                4060,
                4,
                988,
                300,
                436,
                1621,
                294,
                17977,
                11,
                51564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18679775146955854,
            "compression_ratio": 1.5418502202643172,
            "no_speech_prob": 0.062744140625,
            "words": [
                {
                    "start": 1286.02,
                    "end": 1286.4,
                    "word": " Supposing",
                    "probability": 0.917236328125
                },
                {
                    "start": 1286.4,
                    "end": 1287.14,
                    "word": " you're",
                    "probability": 0.89306640625
                },
                {
                    "start": 1287.14,
                    "end": 1287.32,
                    "word": " about",
                    "probability": 0.81494140625
                },
                {
                    "start": 1287.32,
                    "end": 1287.7,
                    "word": " 60",
                    "probability": 0.892578125
                },
                {
                    "start": 1287.7,
                    "end": 1287.94,
                    "word": "%",
                    "probability": 0.90673828125
                },
                {
                    "start": 1287.94,
                    "end": 1288.36,
                    "word": " sure",
                    "probability": 0.787109375
                },
                {
                    "start": 1288.36,
                    "end": 1288.54,
                    "word": " that",
                    "probability": 0.8828125
                },
                {
                    "start": 1288.54,
                    "end": 1288.66,
                    "word": " they",
                    "probability": 0.73974609375
                },
                {
                    "start": 1288.66,
                    "end": 1288.86,
                    "word": " live",
                    "probability": 0.88623046875
                },
                {
                    "start": 1288.86,
                    "end": 1288.98,
                    "word": " in",
                    "probability": 0.90673828125
                },
                {
                    "start": 1288.98,
                    "end": 1289.44,
                    "word": " Wisconsin",
                    "probability": 0.81005859375
                },
                {
                    "start": 1289.44,
                    "end": 1289.9,
                    "word": ",",
                    "probability": 0.4560546875
                }
            ]
        },
        {
            "id": 188,
            "seek": 131383,
            "start": 1289.9,
            "end": 1296.38,
            "text": " and the attacker is attempting to improve that confidence from their perspective, ideally to 100%.",
            "tokens": [
                50364,
                293,
                264,
                35871,
                307,
                22001,
                281,
                3470,
                300,
                6687,
                490,
                641,
                4585,
                11,
                22915,
                281,
                2319,
                6856,
                50664
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20845588235294119,
            "compression_ratio": 1.7008547008547008,
            "no_speech_prob": 0.108154296875,
            "words": [
                {
                    "start": 1289.9,
                    "end": 1290.06,
                    "word": " and",
                    "probability": 0.291748046875
                },
                {
                    "start": 1290.06,
                    "end": 1291.28,
                    "word": " the",
                    "probability": 0.333984375
                },
                {
                    "start": 1291.28,
                    "end": 1291.72,
                    "word": " attacker",
                    "probability": 0.78271484375
                },
                {
                    "start": 1291.72,
                    "end": 1291.94,
                    "word": " is",
                    "probability": 0.88525390625
                },
                {
                    "start": 1291.94,
                    "end": 1292.3,
                    "word": " attempting",
                    "probability": 0.73779296875
                },
                {
                    "start": 1292.3,
                    "end": 1292.6,
                    "word": " to",
                    "probability": 0.90283203125
                },
                {
                    "start": 1292.6,
                    "end": 1293.0,
                    "word": " improve",
                    "probability": 0.73193359375
                },
                {
                    "start": 1293.0,
                    "end": 1293.28,
                    "word": " that",
                    "probability": 0.83740234375
                },
                {
                    "start": 1293.28,
                    "end": 1293.9,
                    "word": " confidence",
                    "probability": 0.8974609375
                },
                {
                    "start": 1293.9,
                    "end": 1294.52,
                    "word": " from",
                    "probability": 0.57861328125
                },
                {
                    "start": 1294.52,
                    "end": 1294.68,
                    "word": " their",
                    "probability": 0.89306640625
                },
                {
                    "start": 1294.68,
                    "end": 1295.08,
                    "word": " perspective",
                    "probability": 0.7578125
                },
                {
                    "start": 1295.08,
                    "end": 1295.18,
                    "word": ",",
                    "probability": 0.50439453125
                },
                {
                    "start": 1295.18,
                    "end": 1295.44,
                    "word": " ideally",
                    "probability": 0.787109375
                },
                {
                    "start": 1295.44,
                    "end": 1295.66,
                    "word": " to",
                    "probability": 0.77392578125
                },
                {
                    "start": 1295.66,
                    "end": 1296.38,
                    "word": " 100%.",
                    "probability": 0.751220703125
                }
            ]
        },
        {
            "id": 189,
            "seek": 131383,
            "start": 1296.88,
            "end": 1304.22,
            "text": " Well, it turns out that there's a really nice translation between how much an attacker can improve their confidence",
            "tokens": [
                50714,
                1042,
                11,
                309,
                4523,
                484,
                300,
                456,
                311,
                257,
                534,
                1481,
                12853,
                1296,
                577,
                709,
                364,
                35871,
                393,
                3470,
                641,
                6687,
                51064
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20845588235294119,
            "compression_ratio": 1.7008547008547008,
            "no_speech_prob": 0.108154296875,
            "words": [
                {
                    "start": 1296.88,
                    "end": 1297.18,
                    "word": " Well",
                    "probability": 0.70654296875
                },
                {
                    "start": 1297.18,
                    "end": 1297.24,
                    "word": ",",
                    "probability": 0.87890625
                },
                {
                    "start": 1297.24,
                    "end": 1297.32,
                    "word": " it",
                    "probability": 0.89501953125
                },
                {
                    "start": 1297.32,
                    "end": 1297.5,
                    "word": " turns",
                    "probability": 0.76708984375
                },
                {
                    "start": 1297.5,
                    "end": 1297.66,
                    "word": " out",
                    "probability": 0.82275390625
                },
                {
                    "start": 1297.66,
                    "end": 1297.78,
                    "word": " that",
                    "probability": 0.857421875
                },
                {
                    "start": 1297.78,
                    "end": 1297.96,
                    "word": " there's",
                    "probability": 0.912109375
                },
                {
                    "start": 1298.16,
                    "end": 1298.48,
                    "word": " a really",
                    "probability": 0.77099609375
                },
                {
                    "start": 1298.48,
                    "end": 1298.96,
                    "word": " nice",
                    "probability": 0.81396484375
                },
                {
                    "start": 1298.96,
                    "end": 1299.48,
                    "word": " translation",
                    "probability": 0.71044921875
                },
                {
                    "start": 1299.48,
                    "end": 1300.08,
                    "word": " between",
                    "probability": 0.77978515625
                },
                {
                    "start": 1300.08,
                    "end": 1301.36,
                    "word": " how",
                    "probability": 0.8896484375
                },
                {
                    "start": 1301.36,
                    "end": 1301.72,
                    "word": " much",
                    "probability": 0.81640625
                },
                {
                    "start": 1301.72,
                    "end": 1301.86,
                    "word": " an",
                    "probability": 0.90380859375
                },
                {
                    "start": 1301.86,
                    "end": 1302.16,
                    "word": " attacker",
                    "probability": 0.70947265625
                },
                {
                    "start": 1302.16,
                    "end": 1302.62,
                    "word": " can",
                    "probability": 0.8994140625
                },
                {
                    "start": 1302.62,
                    "end": 1303.22,
                    "word": " improve",
                    "probability": 0.71435546875
                },
                {
                    "start": 1303.22,
                    "end": 1303.62,
                    "word": " their",
                    "probability": 0.900390625
                },
                {
                    "start": 1303.62,
                    "end": 1304.22,
                    "word": " confidence",
                    "probability": 0.90283203125
                }
            ]
        },
        {
            "id": 190,
            "seek": 131383,
            "start": 1304.22,
            "end": 1310.08,
            "text": " in knowing something and the epsilon that you choose.",
            "tokens": [
                51114,
                294,
                5276,
                746,
                293,
                264,
                17889,
                300,
                291,
                2826,
                13,
                51314
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20845588235294119,
            "compression_ratio": 1.7008547008547008,
            "no_speech_prob": 0.108154296875,
            "words": [
                {
                    "start": 1304.22,
                    "end": 1304.88,
                    "word": " in",
                    "probability": 0.892578125
                },
                {
                    "start": 1304.88,
                    "end": 1307.16,
                    "word": " knowing",
                    "probability": 0.638671875
                },
                {
                    "start": 1307.16,
                    "end": 1307.72,
                    "word": " something",
                    "probability": 0.73779296875
                },
                {
                    "start": 1307.72,
                    "end": 1308.54,
                    "word": " and",
                    "probability": 0.580078125
                },
                {
                    "start": 1308.54,
                    "end": 1308.7,
                    "word": " the",
                    "probability": 0.81005859375
                },
                {
                    "start": 1308.7,
                    "end": 1308.9,
                    "word": " epsilon",
                    "probability": 0.67529296875
                },
                {
                    "start": 1308.9,
                    "end": 1309.12,
                    "word": " that",
                    "probability": 0.85693359375
                },
                {
                    "start": 1309.12,
                    "end": 1309.22,
                    "word": " you",
                    "probability": 0.8974609375
                },
                {
                    "start": 1309.22,
                    "end": 1309.58,
                    "word": " choose",
                    "probability": 0.81982421875
                },
                {
                    "start": 1309.58,
                    "end": 1310.08,
                    "word": ".",
                    "probability": 0.97900390625
                }
            ]
        },
        {
            "id": 191,
            "seek": 131383,
            "start": 1310.08,
            "end": 1318.2,
            "text": " So for a particular choice of epsilon, perhaps they might be able to improve only 5% their confidence if they're starting at 60%.",
            "tokens": [
                51364,
                407,
                337,
                257,
                1729,
                3922,
                295,
                17889,
                11,
                4317,
                436,
                1062,
                312,
                1075,
                281,
                3470,
                787,
                1025,
                4,
                641,
                6687,
                498,
                436,
                434,
                2891,
                412,
                4060,
                6856,
                51764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20845588235294119,
            "compression_ratio": 1.7008547008547008,
            "no_speech_prob": 0.108154296875,
            "words": [
                {
                    "start": 1310.08,
                    "end": 1310.28,
                    "word": " So",
                    "probability": 0.79638671875
                },
                {
                    "start": 1310.28,
                    "end": 1310.7,
                    "word": " for",
                    "probability": 0.66796875
                },
                {
                    "start": 1310.92,
                    "end": 1311.1,
                    "word": " a particular",
                    "probability": 0.8251953125
                },
                {
                    "start": 1311.1,
                    "end": 1311.46,
                    "word": " choice",
                    "probability": 0.81298828125
                },
                {
                    "start": 1311.46,
                    "end": 1311.58,
                    "word": " of",
                    "probability": 0.9013671875
                },
                {
                    "start": 1311.58,
                    "end": 1311.86,
                    "word": " epsilon",
                    "probability": 0.75146484375
                },
                {
                    "start": 1311.86,
                    "end": 1312.12,
                    "word": ",",
                    "probability": 0.93017578125
                },
                {
                    "start": 1312.12,
                    "end": 1312.34,
                    "word": " perhaps",
                    "probability": 0.82373046875
                },
                {
                    "start": 1312.34,
                    "end": 1312.5,
                    "word": " they",
                    "probability": 0.732421875
                },
                {
                    "start": 1312.5,
                    "end": 1312.64,
                    "word": " might",
                    "probability": 0.81787109375
                },
                {
                    "start": 1312.64,
                    "end": 1312.78,
                    "word": " be",
                    "probability": 0.90380859375
                },
                {
                    "start": 1312.78,
                    "end": 1312.92,
                    "word": " able",
                    "probability": 0.8916015625
                },
                {
                    "start": 1312.92,
                    "end": 1313.1,
                    "word": " to",
                    "probability": 0.90673828125
                },
                {
                    "start": 1313.1,
                    "end": 1313.66,
                    "word": " improve",
                    "probability": 0.7138671875
                },
                {
                    "start": 1313.66,
                    "end": 1313.94,
                    "word": " only",
                    "probability": 0.849609375
                },
                {
                    "start": 1313.94,
                    "end": 1314.34,
                    "word": " 5",
                    "probability": 0.9189453125
                },
                {
                    "start": 1314.34,
                    "end": 1314.64,
                    "word": "%",
                    "probability": 0.9833984375
                },
                {
                    "start": 1314.64,
                    "end": 1314.94,
                    "word": " their",
                    "probability": 0.5556640625
                },
                {
                    "start": 1314.94,
                    "end": 1315.48,
                    "word": " confidence",
                    "probability": 0.90625
                },
                {
                    "start": 1315.48,
                    "end": 1316.64,
                    "word": " if",
                    "probability": 0.591796875
                },
                {
                    "start": 1316.64,
                    "end": 1316.86,
                    "word": " they're",
                    "probability": 0.8291015625
                },
                {
                    "start": 1316.86,
                    "end": 1317.2,
                    "word": " starting",
                    "probability": 0.84228515625
                },
                {
                    "start": 1317.2,
                    "end": 1317.36,
                    "word": " at",
                    "probability": 0.90380859375
                },
                {
                    "start": 1317.36,
                    "end": 1318.2,
                    "word": " 60%.",
                    "probability": 0.9208984375
                }
            ]
        },
        {
            "id": 192,
            "seek": 133991,
            "start": 1318.2,
            "end": 1325.92,
            "text": " If you really, really want to be rigorous, perhaps they're not able to really improve it, you know, in any sort of measurable way at all.",
            "tokens": [
                50364,
                759,
                291,
                534,
                11,
                534,
                528,
                281,
                312,
                29882,
                11,
                4317,
                436,
                434,
                406,
                1075,
                281,
                534,
                3470,
                309,
                11,
                291,
                458,
                11,
                294,
                604,
                1333,
                295,
                43615,
                636,
                412,
                439,
                13,
                50714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2094089724123478,
            "compression_ratio": 1.6099585062240664,
            "no_speech_prob": 0.0094146728515625,
            "words": [
                {
                    "start": 1318.2,
                    "end": 1319.02,
                    "word": " If",
                    "probability": 0.52587890625
                },
                {
                    "start": 1319.02,
                    "end": 1319.16,
                    "word": " you",
                    "probability": 0.8955078125
                },
                {
                    "start": 1319.16,
                    "end": 1319.48,
                    "word": " really",
                    "probability": 0.77490234375
                },
                {
                    "start": 1319.48,
                    "end": 1319.58,
                    "word": ",",
                    "probability": 0.70751953125
                },
                {
                    "start": 1319.58,
                    "end": 1319.68,
                    "word": " really",
                    "probability": 0.77001953125
                },
                {
                    "start": 1319.68,
                    "end": 1319.88,
                    "word": " want",
                    "probability": 0.728515625
                },
                {
                    "start": 1319.88,
                    "end": 1319.96,
                    "word": " to",
                    "probability": 0.90625
                },
                {
                    "start": 1319.96,
                    "end": 1320.08,
                    "word": " be",
                    "probability": 0.904296875
                },
                {
                    "start": 1320.08,
                    "end": 1320.44,
                    "word": " rigorous",
                    "probability": 0.84912109375
                },
                {
                    "start": 1320.44,
                    "end": 1320.7,
                    "word": ",",
                    "probability": 0.8828125
                },
                {
                    "start": 1320.7,
                    "end": 1321.16,
                    "word": " perhaps",
                    "probability": 0.8212890625
                },
                {
                    "start": 1321.16,
                    "end": 1321.38,
                    "word": " they're",
                    "probability": 0.83349609375
                },
                {
                    "start": 1321.38,
                    "end": 1321.52,
                    "word": " not",
                    "probability": 0.8916015625
                },
                {
                    "start": 1321.52,
                    "end": 1321.72,
                    "word": " able",
                    "probability": 0.8876953125
                },
                {
                    "start": 1321.72,
                    "end": 1321.92,
                    "word": " to",
                    "probability": 0.904296875
                },
                {
                    "start": 1321.92,
                    "end": 1322.22,
                    "word": " really",
                    "probability": 0.79345703125
                },
                {
                    "start": 1322.22,
                    "end": 1322.68,
                    "word": " improve",
                    "probability": 0.71630859375
                },
                {
                    "start": 1322.68,
                    "end": 1322.9,
                    "word": " it",
                    "probability": 0.90478515625
                },
                {
                    "start": 1322.9,
                    "end": 1322.94,
                    "word": ",",
                    "probability": 0.41650390625
                },
                {
                    "start": 1322.94,
                    "end": 1323.44,
                    "word": " you",
                    "probability": 0.76806640625
                },
                {
                    "start": 1323.44,
                    "end": 1323.58,
                    "word": " know",
                    "probability": 0.8212890625
                },
                {
                    "start": 1323.58,
                    "end": 1323.6,
                    "word": ",",
                    "probability": 0.9365234375
                },
                {
                    "start": 1323.6,
                    "end": 1323.74,
                    "word": " in",
                    "probability": 0.90380859375
                },
                {
                    "start": 1323.74,
                    "end": 1324.1,
                    "word": " any",
                    "probability": 0.7939453125
                },
                {
                    "start": 1324.1,
                    "end": 1324.28,
                    "word": " sort",
                    "probability": 0.8310546875
                },
                {
                    "start": 1324.28,
                    "end": 1324.38,
                    "word": " of",
                    "probability": 0.904296875
                },
                {
                    "start": 1324.38,
                    "end": 1324.66,
                    "word": " measurable",
                    "probability": 0.83740234375
                },
                {
                    "start": 1324.66,
                    "end": 1325.04,
                    "word": " way",
                    "probability": 0.90234375
                },
                {
                    "start": 1325.04,
                    "end": 1325.2,
                    "word": " at",
                    "probability": 0.9013671875
                },
                {
                    "start": 1325.2,
                    "end": 1325.46,
                    "word": " all",
                    "probability": 0.8974609375
                },
                {
                    "start": 1325.46,
                    "end": 1325.92,
                    "word": ".",
                    "probability": 0.9462890625
                }
            ]
        },
        {
            "id": 193,
            "seek": 133991,
            "start": 1325.92,
            "end": 1330.74,
            "text": " And so you have this nice principled trade-off between risk and accuracy,",
            "tokens": [
                50764,
                400,
                370,
                291,
                362,
                341,
                1481,
                3681,
                15551,
                4923,
                12,
                4506,
                1296,
                3148,
                293,
                14170,
                11,
                50964
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2094089724123478,
            "compression_ratio": 1.6099585062240664,
            "no_speech_prob": 0.0094146728515625,
            "words": [
                {
                    "start": 1325.92,
                    "end": 1326.2,
                    "word": " And",
                    "probability": 0.875
                },
                {
                    "start": 1326.2,
                    "end": 1326.34,
                    "word": " so",
                    "probability": 0.9013671875
                },
                {
                    "start": 1326.34,
                    "end": 1326.46,
                    "word": " you",
                    "probability": 0.69287109375
                },
                {
                    "start": 1326.46,
                    "end": 1326.84,
                    "word": " have",
                    "probability": 0.88623046875
                },
                {
                    "start": 1326.84,
                    "end": 1327.16,
                    "word": " this",
                    "probability": 0.88330078125
                },
                {
                    "start": 1327.16,
                    "end": 1327.46,
                    "word": " nice",
                    "probability": 0.8017578125
                },
                {
                    "start": 1327.46,
                    "end": 1327.98,
                    "word": " principled",
                    "probability": 0.646728515625
                },
                {
                    "start": 1327.98,
                    "end": 1328.26,
                    "word": " trade",
                    "probability": 0.841796875
                },
                {
                    "start": 1328.26,
                    "end": 1328.46,
                    "word": "-off",
                    "probability": 0.6278076171875
                },
                {
                    "start": 1328.46,
                    "end": 1328.72,
                    "word": " between",
                    "probability": 0.775390625
                },
                {
                    "start": 1328.72,
                    "end": 1329.18,
                    "word": " risk",
                    "probability": 0.80615234375
                },
                {
                    "start": 1329.18,
                    "end": 1329.94,
                    "word": " and",
                    "probability": 0.88232421875
                },
                {
                    "start": 1329.94,
                    "end": 1330.48,
                    "word": " accuracy",
                    "probability": 0.75
                },
                {
                    "start": 1330.48,
                    "end": 1330.74,
                    "word": ",",
                    "probability": 0.495361328125
                }
            ]
        },
        {
            "id": 194,
            "seek": 133991,
            "start": 1330.74,
            "end": 1339.16,
            "text": " since the epsilon has the strongest influence on sort of the quality, the accuracy, the utility of the output results.",
            "tokens": [
                51014,
                1670,
                264,
                17889,
                575,
                264,
                16595,
                6503,
                322,
                1333,
                295,
                264,
                3125,
                11,
                264,
                14170,
                11,
                264,
                14877,
                295,
                264,
                5598,
                3542,
                13,
                51364
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2094089724123478,
            "compression_ratio": 1.6099585062240664,
            "no_speech_prob": 0.0094146728515625,
            "words": [
                {
                    "start": 1330.74,
                    "end": 1330.98,
                    "word": " since",
                    "probability": 0.78564453125
                },
                {
                    "start": 1330.98,
                    "end": 1331.16,
                    "word": " the",
                    "probability": 0.81689453125
                },
                {
                    "start": 1331.16,
                    "end": 1331.4,
                    "word": " epsilon",
                    "probability": 0.70556640625
                },
                {
                    "start": 1331.4,
                    "end": 1332.12,
                    "word": " has",
                    "probability": 0.90576171875
                },
                {
                    "start": 1332.12,
                    "end": 1332.44,
                    "word": " the",
                    "probability": 0.8134765625
                },
                {
                    "start": 1332.44,
                    "end": 1332.9,
                    "word": " strongest",
                    "probability": 0.84619140625
                },
                {
                    "start": 1332.9,
                    "end": 1334.22,
                    "word": " influence",
                    "probability": 0.8330078125
                },
                {
                    "start": 1334.22,
                    "end": 1334.5,
                    "word": " on",
                    "probability": 0.90185546875
                },
                {
                    "start": 1334.5,
                    "end": 1334.7,
                    "word": " sort",
                    "probability": 0.69873046875
                },
                {
                    "start": 1334.7,
                    "end": 1334.8,
                    "word": " of",
                    "probability": 0.9033203125
                },
                {
                    "start": 1334.8,
                    "end": 1335.08,
                    "word": " the",
                    "probability": 0.8076171875
                },
                {
                    "start": 1335.08,
                    "end": 1336.36,
                    "word": " quality",
                    "probability": 0.62451171875
                },
                {
                    "start": 1336.36,
                    "end": 1336.58,
                    "word": ",",
                    "probability": 0.93115234375
                },
                {
                    "start": 1336.58,
                    "end": 1336.68,
                    "word": " the",
                    "probability": 0.80126953125
                },
                {
                    "start": 1336.68,
                    "end": 1337.06,
                    "word": " accuracy",
                    "probability": 0.751953125
                },
                {
                    "start": 1337.06,
                    "end": 1337.2,
                    "word": ",",
                    "probability": 0.8798828125
                },
                {
                    "start": 1337.2,
                    "end": 1337.28,
                    "word": " the",
                    "probability": 0.76513671875
                },
                {
                    "start": 1337.28,
                    "end": 1337.62,
                    "word": " utility",
                    "probability": 0.8095703125
                },
                {
                    "start": 1337.62,
                    "end": 1337.8,
                    "word": " of",
                    "probability": 0.884765625
                },
                {
                    "start": 1337.8,
                    "end": 1337.9,
                    "word": " the",
                    "probability": 0.81103515625
                },
                {
                    "start": 1337.9,
                    "end": 1338.12,
                    "word": " output",
                    "probability": 0.8271484375
                },
                {
                    "start": 1338.12,
                    "end": 1338.64,
                    "word": " results",
                    "probability": 0.71484375
                },
                {
                    "start": 1338.64,
                    "end": 1339.16,
                    "word": ".",
                    "probability": 0.9921875
                }
            ]
        },
        {
            "id": 195,
            "seek": 133991,
            "start": 1340.48,
            "end": 1344.28,
            "text": " Delta is a much easier concept to understand and explain.",
            "tokens": [
                51414,
                18183,
                307,
                257,
                709,
                3571,
                3410,
                281,
                1223,
                293,
                2903,
                13,
                51614
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2094089724123478,
            "compression_ratio": 1.6099585062240664,
            "no_speech_prob": 0.0094146728515625,
            "words": [
                {
                    "start": 1340.48,
                    "end": 1341.0,
                    "word": " Delta",
                    "probability": 0.84619140625
                },
                {
                    "start": 1341.0,
                    "end": 1341.16,
                    "word": " is",
                    "probability": 0.748046875
                },
                {
                    "start": 1341.34,
                    "end": 1341.56,
                    "word": " a much",
                    "probability": 0.81640625
                },
                {
                    "start": 1341.56,
                    "end": 1341.94,
                    "word": " easier",
                    "probability": 0.921875
                },
                {
                    "start": 1341.94,
                    "end": 1342.44,
                    "word": " concept",
                    "probability": 0.7861328125
                },
                {
                    "start": 1342.44,
                    "end": 1342.62,
                    "word": " to",
                    "probability": 0.904296875
                },
                {
                    "start": 1342.62,
                    "end": 1343.06,
                    "word": " understand",
                    "probability": 0.744140625
                },
                {
                    "start": 1343.06,
                    "end": 1343.32,
                    "word": " and",
                    "probability": 0.89013671875
                },
                {
                    "start": 1343.32,
                    "end": 1343.76,
                    "word": " explain",
                    "probability": 0.8408203125
                },
                {
                    "start": 1343.76,
                    "end": 1344.28,
                    "word": ".",
                    "probability": 0.96728515625
                }
            ]
        },
        {
            "id": 196,
            "seek": 136837,
            "start": 1344.28,
            "end": 1351.58,
            "text": " It's basically our out to say that we basically made a differentially private engine,",
            "tokens": [
                50364,
                467,
                311,
                1936,
                527,
                484,
                281,
                584,
                300,
                321,
                1936,
                1027,
                257,
                819,
                2270,
                4551,
                2848,
                11,
                50714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.16775568683039058,
            "compression_ratio": 1.8255813953488371,
            "no_speech_prob": 0.048126220703125,
            "words": [
                {
                    "start": 1344.28,
                    "end": 1344.64,
                    "word": " It's",
                    "probability": 0.651123046875
                },
                {
                    "start": 1344.64,
                    "end": 1345.0,
                    "word": " basically",
                    "probability": 0.90673828125
                },
                {
                    "start": 1345.0,
                    "end": 1345.3,
                    "word": " our",
                    "probability": 0.4736328125
                },
                {
                    "start": 1345.3,
                    "end": 1345.7,
                    "word": " out",
                    "probability": 0.79345703125
                },
                {
                    "start": 1345.7,
                    "end": 1346.0,
                    "word": " to",
                    "probability": 0.8583984375
                },
                {
                    "start": 1346.0,
                    "end": 1346.24,
                    "word": " say",
                    "probability": 0.896484375
                },
                {
                    "start": 1346.24,
                    "end": 1346.5,
                    "word": " that",
                    "probability": 0.869140625
                },
                {
                    "start": 1346.5,
                    "end": 1346.66,
                    "word": " we",
                    "probability": 0.89453125
                },
                {
                    "start": 1346.66,
                    "end": 1347.16,
                    "word": " basically",
                    "probability": 0.86669921875
                },
                {
                    "start": 1347.16,
                    "end": 1347.56,
                    "word": " made",
                    "probability": 0.81591796875
                },
                {
                    "start": 1348.86,
                    "end": 1349.3,
                    "word": " a differentially",
                    "probability": 0.77001953125
                },
                {
                    "start": 1349.3,
                    "end": 1349.66,
                    "word": " private",
                    "probability": 0.7578125
                },
                {
                    "start": 1349.66,
                    "end": 1351.3,
                    "word": " engine",
                    "probability": 0.64794921875
                },
                {
                    "start": 1351.3,
                    "end": 1351.58,
                    "word": ",",
                    "probability": 0.449951171875
                }
            ]
        },
        {
            "id": 197,
            "seek": 136837,
            "start": 1351.58,
            "end": 1356.02,
            "text": " but occasionally, very occasionally, it's not differentially private.",
            "tokens": [
                50764,
                457,
                16895,
                11,
                588,
                16895,
                11,
                309,
                311,
                406,
                819,
                2270,
                4551,
                13,
                50914
            ],
            "temperature": 0.0,
            "avg_logprob": -0.16775568683039058,
            "compression_ratio": 1.8255813953488371,
            "no_speech_prob": 0.048126220703125,
            "words": [
                {
                    "start": 1351.58,
                    "end": 1352.18,
                    "word": " but",
                    "probability": 0.8935546875
                },
                {
                    "start": 1352.18,
                    "end": 1352.8,
                    "word": " occasionally",
                    "probability": 0.7177734375
                },
                {
                    "start": 1352.8,
                    "end": 1353.22,
                    "word": ",",
                    "probability": 0.79443359375
                },
                {
                    "start": 1353.22,
                    "end": 1353.42,
                    "word": " very",
                    "probability": 0.72705078125
                },
                {
                    "start": 1353.42,
                    "end": 1353.94,
                    "word": " occasionally",
                    "probability": 0.845703125
                },
                {
                    "start": 1353.94,
                    "end": 1354.3,
                    "word": ",",
                    "probability": 0.79443359375
                },
                {
                    "start": 1354.3,
                    "end": 1354.56,
                    "word": " it's",
                    "probability": 0.9482421875
                },
                {
                    "start": 1354.56,
                    "end": 1354.74,
                    "word": " not",
                    "probability": 0.9052734375
                },
                {
                    "start": 1354.74,
                    "end": 1355.18,
                    "word": " differentially",
                    "probability": 0.87158203125
                },
                {
                    "start": 1355.18,
                    "end": 1355.5,
                    "word": " private",
                    "probability": 0.7880859375
                },
                {
                    "start": 1355.5,
                    "end": 1356.02,
                    "word": ".",
                    "probability": 0.9677734375
                }
            ]
        },
        {
            "id": 198,
            "seek": 136837,
            "start": 1356.02,
            "end": 1358.94,
            "text": " So delta is the probability of catastrophic failure.",
            "tokens": [
                50964,
                407,
                8289,
                307,
                264,
                8482,
                295,
                34915,
                7763,
                13,
                51064
            ],
            "temperature": 0.0,
            "avg_logprob": -0.16775568683039058,
            "compression_ratio": 1.8255813953488371,
            "no_speech_prob": 0.048126220703125,
            "words": [
                {
                    "start": 1356.02,
                    "end": 1356.12,
                    "word": " So",
                    "probability": 0.8359375
                },
                {
                    "start": 1356.12,
                    "end": 1356.34,
                    "word": " delta",
                    "probability": 0.43408203125
                },
                {
                    "start": 1356.34,
                    "end": 1356.58,
                    "word": " is",
                    "probability": 0.908203125
                },
                {
                    "start": 1356.58,
                    "end": 1356.72,
                    "word": " the",
                    "probability": 0.81689453125
                },
                {
                    "start": 1356.72,
                    "end": 1357.12,
                    "word": " probability",
                    "probability": 0.89111328125
                },
                {
                    "start": 1357.12,
                    "end": 1357.38,
                    "word": " of",
                    "probability": 0.9091796875
                },
                {
                    "start": 1357.38,
                    "end": 1357.86,
                    "word": " catastrophic",
                    "probability": 0.736328125
                },
                {
                    "start": 1357.86,
                    "end": 1358.42,
                    "word": " failure",
                    "probability": 0.81689453125
                },
                {
                    "start": 1358.42,
                    "end": 1358.94,
                    "word": ".",
                    "probability": 0.984375
                }
            ]
        },
        {
            "id": 199,
            "seek": 136837,
            "start": 1359.16,
            "end": 1365.32,
            "text": " There are lots of ways to think about it, but the way we generally talk about it is perhaps the number of output rows",
            "tokens": [
                51114,
                821,
                366,
                3195,
                295,
                2098,
                281,
                519,
                466,
                309,
                11,
                457,
                264,
                636,
                321,
                5101,
                751,
                466,
                309,
                307,
                4317,
                264,
                1230,
                295,
                5598,
                13241,
                51414
            ],
            "temperature": 0.0,
            "avg_logprob": -0.16775568683039058,
            "compression_ratio": 1.8255813953488371,
            "no_speech_prob": 0.048126220703125,
            "words": [
                {
                    "start": 1359.16,
                    "end": 1359.5,
                    "word": " There",
                    "probability": 0.791015625
                },
                {
                    "start": 1359.5,
                    "end": 1359.58,
                    "word": " are",
                    "probability": 0.8876953125
                },
                {
                    "start": 1359.58,
                    "end": 1359.76,
                    "word": " lots",
                    "probability": 0.80517578125
                },
                {
                    "start": 1359.76,
                    "end": 1359.84,
                    "word": " of",
                    "probability": 0.904296875
                },
                {
                    "start": 1359.84,
                    "end": 1359.98,
                    "word": " ways",
                    "probability": 0.80517578125
                },
                {
                    "start": 1359.98,
                    "end": 1360.1,
                    "word": " to",
                    "probability": 0.90380859375
                },
                {
                    "start": 1360.1,
                    "end": 1360.26,
                    "word": " think",
                    "probability": 0.8095703125
                },
                {
                    "start": 1360.26,
                    "end": 1360.52,
                    "word": " about",
                    "probability": 0.82275390625
                },
                {
                    "start": 1360.52,
                    "end": 1360.7,
                    "word": " it",
                    "probability": 0.90283203125
                },
                {
                    "start": 1360.7,
                    "end": 1360.8,
                    "word": ",",
                    "probability": 0.748046875
                },
                {
                    "start": 1360.8,
                    "end": 1360.96,
                    "word": " but",
                    "probability": 0.89404296875
                },
                {
                    "start": 1360.96,
                    "end": 1361.18,
                    "word": " the",
                    "probability": 0.814453125
                },
                {
                    "start": 1361.18,
                    "end": 1361.3,
                    "word": " way",
                    "probability": 0.904296875
                },
                {
                    "start": 1361.3,
                    "end": 1361.44,
                    "word": " we",
                    "probability": 0.90283203125
                },
                {
                    "start": 1361.44,
                    "end": 1361.88,
                    "word": " generally",
                    "probability": 0.87939453125
                },
                {
                    "start": 1361.88,
                    "end": 1362.18,
                    "word": " talk",
                    "probability": 0.8056640625
                },
                {
                    "start": 1362.18,
                    "end": 1362.4,
                    "word": " about",
                    "probability": 0.82470703125
                },
                {
                    "start": 1362.4,
                    "end": 1362.54,
                    "word": " it",
                    "probability": 0.9033203125
                },
                {
                    "start": 1362.54,
                    "end": 1362.74,
                    "word": " is",
                    "probability": 0.896484375
                },
                {
                    "start": 1362.74,
                    "end": 1363.18,
                    "word": " perhaps",
                    "probability": 0.7421875
                },
                {
                    "start": 1363.18,
                    "end": 1363.4,
                    "word": " the",
                    "probability": 0.8125
                },
                {
                    "start": 1363.4,
                    "end": 1363.76,
                    "word": " number",
                    "probability": 0.8798828125
                },
                {
                    "start": 1363.76,
                    "end": 1364.34,
                    "word": " of",
                    "probability": 0.904296875
                },
                {
                    "start": 1364.34,
                    "end": 1364.92,
                    "word": " output",
                    "probability": 0.8271484375
                },
                {
                    "start": 1364.92,
                    "end": 1365.32,
                    "word": " rows",
                    "probability": 0.8203125
                }
            ]
        },
        {
            "id": 200,
            "seek": 136837,
            "start": 1365.32,
            "end": 1368.74,
            "text": " that are not strictly differentially private.",
            "tokens": [
                51464,
                300,
                366,
                406,
                20792,
                819,
                2270,
                4551,
                13,
                51564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.16775568683039058,
            "compression_ratio": 1.8255813953488371,
            "no_speech_prob": 0.048126220703125,
            "words": [
                {
                    "start": 1365.32,
                    "end": 1366.18,
                    "word": " that",
                    "probability": 0.869140625
                },
                {
                    "start": 1366.18,
                    "end": 1366.28,
                    "word": " are",
                    "probability": 0.9013671875
                },
                {
                    "start": 1366.28,
                    "end": 1366.56,
                    "word": " not",
                    "probability": 0.89892578125
                },
                {
                    "start": 1366.56,
                    "end": 1367.16,
                    "word": " strictly",
                    "probability": 0.83251953125
                },
                {
                    "start": 1367.16,
                    "end": 1368.06,
                    "word": " differentially",
                    "probability": 0.880126953125
                },
                {
                    "start": 1368.06,
                    "end": 1368.38,
                    "word": " private",
                    "probability": 0.7978515625
                },
                {
                    "start": 1368.38,
                    "end": 1368.74,
                    "word": ".",
                    "probability": 0.98291015625
                }
            ]
        },
        {
            "id": 201,
            "seek": 136837,
            "start": 1368.74,
            "end": 1372.74,
            "text": " Now, of course, you don't know which rows those are, so there's always sort of a reasonable doubt.",
            "tokens": [
                51614,
                823,
                11,
                295,
                1164,
                11,
                291,
                500,
                380,
                458,
                597,
                13241,
                729,
                366,
                11,
                370,
                456,
                311,
                1009,
                1333,
                295,
                257,
                10585,
                6385,
                13,
                51764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.16775568683039058,
            "compression_ratio": 1.8255813953488371,
            "no_speech_prob": 0.048126220703125,
            "words": [
                {
                    "start": 1368.74,
                    "end": 1368.84,
                    "word": " Now",
                    "probability": 0.69921875
                },
                {
                    "start": 1368.84,
                    "end": 1368.88,
                    "word": ",",
                    "probability": 0.71044921875
                },
                {
                    "start": 1368.88,
                    "end": 1368.94,
                    "word": " of",
                    "probability": 0.90625
                },
                {
                    "start": 1368.94,
                    "end": 1369.06,
                    "word": " course",
                    "probability": 0.8662109375
                },
                {
                    "start": 1369.06,
                    "end": 1369.1,
                    "word": ",",
                    "probability": 0.87451171875
                },
                {
                    "start": 1369.1,
                    "end": 1369.18,
                    "word": " you",
                    "probability": 0.89111328125
                },
                {
                    "start": 1369.18,
                    "end": 1369.34,
                    "word": " don't",
                    "probability": 0.946044921875
                },
                {
                    "start": 1369.34,
                    "end": 1369.54,
                    "word": " know",
                    "probability": 0.806640625
                },
                {
                    "start": 1369.54,
                    "end": 1369.72,
                    "word": " which",
                    "probability": 0.9052734375
                },
                {
                    "start": 1369.72,
                    "end": 1369.96,
                    "word": " rows",
                    "probability": 0.8037109375
                },
                {
                    "start": 1369.96,
                    "end": 1370.16,
                    "word": " those",
                    "probability": 0.841796875
                },
                {
                    "start": 1370.16,
                    "end": 1370.46,
                    "word": " are",
                    "probability": 0.8916015625
                },
                {
                    "start": 1370.46,
                    "end": 1370.54,
                    "word": ",",
                    "probability": 0.66796875
                },
                {
                    "start": 1370.54,
                    "end": 1370.72,
                    "word": " so",
                    "probability": 0.8984375
                },
                {
                    "start": 1370.72,
                    "end": 1370.94,
                    "word": " there's",
                    "probability": 0.92236328125
                },
                {
                    "start": 1370.94,
                    "end": 1371.16,
                    "word": " always",
                    "probability": 0.7705078125
                },
                {
                    "start": 1371.16,
                    "end": 1371.38,
                    "word": " sort",
                    "probability": 0.68603515625
                },
                {
                    "start": 1371.38,
                    "end": 1371.44,
                    "word": " of",
                    "probability": 0.9013671875
                },
                {
                    "start": 1371.54,
                    "end": 1371.82,
                    "word": " a reasonable",
                    "probability": 0.802734375
                },
                {
                    "start": 1371.82,
                    "end": 1372.22,
                    "word": " doubt",
                    "probability": 0.86865234375
                },
                {
                    "start": 1372.22,
                    "end": 1372.74,
                    "word": ".",
                    "probability": 0.30126953125
                }
            ]
        },
        {
            "id": 202,
            "seek": 139397,
            "start": 1372.74,
            "end": 1381.54,
            "text": " But this allows you to really quantify a baseline of failure modes that your output data set can have.",
            "tokens": [
                50364,
                583,
                341,
                4045,
                291,
                281,
                534,
                40421,
                257,
                20518,
                295,
                7763,
                14068,
                300,
                428,
                5598,
                1412,
                992,
                393,
                362,
                13,
                50764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2133333384990692,
            "compression_ratio": 1.528205128205128,
            "no_speech_prob": 0.008697509765625,
            "words": [
                {
                    "start": 1372.74,
                    "end": 1373.22,
                    "word": " But",
                    "probability": 0.6220703125
                },
                {
                    "start": 1373.22,
                    "end": 1373.66,
                    "word": " this",
                    "probability": 0.8583984375
                },
                {
                    "start": 1373.66,
                    "end": 1373.98,
                    "word": " allows",
                    "probability": 0.7890625
                },
                {
                    "start": 1373.98,
                    "end": 1374.12,
                    "word": " you",
                    "probability": 0.89599609375
                },
                {
                    "start": 1374.12,
                    "end": 1374.3,
                    "word": " to",
                    "probability": 0.9091796875
                },
                {
                    "start": 1374.3,
                    "end": 1374.5,
                    "word": " really",
                    "probability": 0.77880859375
                },
                {
                    "start": 1374.5,
                    "end": 1375.18,
                    "word": " quantify",
                    "probability": 0.93115234375
                },
                {
                    "start": 1375.46,
                    "end": 1376.02,
                    "word": " a baseline",
                    "probability": 0.927734375
                },
                {
                    "start": 1376.02,
                    "end": 1377.08,
                    "word": " of",
                    "probability": 0.880859375
                },
                {
                    "start": 1377.08,
                    "end": 1378.58,
                    "word": " failure",
                    "probability": 0.6865234375
                },
                {
                    "start": 1378.58,
                    "end": 1379.24,
                    "word": " modes",
                    "probability": 0.80859375
                },
                {
                    "start": 1379.24,
                    "end": 1379.48,
                    "word": " that",
                    "probability": 0.87646484375
                },
                {
                    "start": 1379.48,
                    "end": 1379.74,
                    "word": " your",
                    "probability": 0.78662109375
                },
                {
                    "start": 1379.74,
                    "end": 1380.12,
                    "word": " output",
                    "probability": 0.86376953125
                },
                {
                    "start": 1380.12,
                    "end": 1380.36,
                    "word": " data",
                    "probability": 0.450439453125
                },
                {
                    "start": 1380.36,
                    "end": 1380.58,
                    "word": " set",
                    "probability": 0.84912109375
                },
                {
                    "start": 1380.58,
                    "end": 1380.7,
                    "word": " can",
                    "probability": 0.89794921875
                },
                {
                    "start": 1380.7,
                    "end": 1381.04,
                    "word": " have",
                    "probability": 0.8837890625
                },
                {
                    "start": 1381.04,
                    "end": 1381.54,
                    "word": ".",
                    "probability": 0.955078125
                }
            ]
        },
        {
            "id": 203,
            "seek": 139397,
            "start": 1381.82,
            "end": 1388.88,
            "text": " Delta tends to be very, very small, typically on the order of 1 over n, where n is the number of users, or rows, rather.",
            "tokens": [
                50814,
                18183,
                12258,
                281,
                312,
                588,
                11,
                588,
                1359,
                11,
                5850,
                322,
                264,
                1668,
                295,
                502,
                670,
                297,
                11,
                689,
                297,
                307,
                264,
                1230,
                295,
                5022,
                11,
                420,
                13241,
                11,
                2831,
                13,
                51114
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2133333384990692,
            "compression_ratio": 1.528205128205128,
            "no_speech_prob": 0.008697509765625,
            "words": [
                {
                    "start": 1381.82,
                    "end": 1381.94,
                    "word": " Delta",
                    "probability": 0.888671875
                },
                {
                    "start": 1381.94,
                    "end": 1382.2,
                    "word": " tends",
                    "probability": 0.71240234375
                },
                {
                    "start": 1382.2,
                    "end": 1382.34,
                    "word": " to",
                    "probability": 0.904296875
                },
                {
                    "start": 1382.34,
                    "end": 1382.48,
                    "word": " be",
                    "probability": 0.9091796875
                },
                {
                    "start": 1382.48,
                    "end": 1382.82,
                    "word": " very",
                    "probability": 0.744140625
                },
                {
                    "start": 1382.82,
                    "end": 1382.86,
                    "word": ",",
                    "probability": 0.87646484375
                },
                {
                    "start": 1382.86,
                    "end": 1383.1,
                    "word": " very",
                    "probability": 0.744140625
                },
                {
                    "start": 1383.1,
                    "end": 1383.56,
                    "word": " small",
                    "probability": 0.84228515625
                },
                {
                    "start": 1383.56,
                    "end": 1384.1,
                    "word": ",",
                    "probability": 0.68359375
                },
                {
                    "start": 1384.1,
                    "end": 1385.04,
                    "word": " typically",
                    "probability": 0.84375
                },
                {
                    "start": 1385.04,
                    "end": 1385.28,
                    "word": " on",
                    "probability": 0.87353515625
                },
                {
                    "start": 1385.28,
                    "end": 1385.38,
                    "word": " the",
                    "probability": 0.81884765625
                },
                {
                    "start": 1385.38,
                    "end": 1385.58,
                    "word": " order",
                    "probability": 0.8095703125
                },
                {
                    "start": 1385.58,
                    "end": 1385.76,
                    "word": " of",
                    "probability": 0.9033203125
                },
                {
                    "start": 1385.76,
                    "end": 1385.92,
                    "word": " 1",
                    "probability": 0.23486328125
                },
                {
                    "start": 1385.92,
                    "end": 1386.16,
                    "word": " over",
                    "probability": 0.54150390625
                },
                {
                    "start": 1386.16,
                    "end": 1386.42,
                    "word": " n",
                    "probability": 0.5751953125
                },
                {
                    "start": 1386.42,
                    "end": 1386.48,
                    "word": ",",
                    "probability": 0.6337890625
                },
                {
                    "start": 1386.48,
                    "end": 1386.58,
                    "word": " where",
                    "probability": 0.90234375
                },
                {
                    "start": 1386.58,
                    "end": 1386.74,
                    "word": " n",
                    "probability": 0.99267578125
                },
                {
                    "start": 1386.74,
                    "end": 1386.86,
                    "word": " is",
                    "probability": 0.9033203125
                },
                {
                    "start": 1386.86,
                    "end": 1386.96,
                    "word": " the",
                    "probability": 0.80908203125
                },
                {
                    "start": 1386.96,
                    "end": 1387.16,
                    "word": " number",
                    "probability": 0.8701171875
                },
                {
                    "start": 1387.16,
                    "end": 1387.32,
                    "word": " of",
                    "probability": 0.90625
                },
                {
                    "start": 1387.32,
                    "end": 1387.64,
                    "word": " users",
                    "probability": 0.83642578125
                },
                {
                    "start": 1387.64,
                    "end": 1387.74,
                    "word": ",",
                    "probability": 0.39697265625
                },
                {
                    "start": 1387.74,
                    "end": 1387.86,
                    "word": " or",
                    "probability": 0.8955078125
                },
                {
                    "start": 1387.86,
                    "end": 1388.12,
                    "word": " rows",
                    "probability": 0.8203125
                },
                {
                    "start": 1388.12,
                    "end": 1388.24,
                    "word": ",",
                    "probability": 0.388916015625
                },
                {
                    "start": 1388.24,
                    "end": 1388.38,
                    "word": " rather",
                    "probability": 0.80712890625
                },
                {
                    "start": 1388.38,
                    "end": 1388.88,
                    "word": ".",
                    "probability": 0.916015625
                }
            ]
        },
        {
            "id": 204,
            "seek": 139397,
            "start": 1390.88,
            "end": 1398.34,
            "text": " And because the nature of the noise that we use is a decaying exponential,",
            "tokens": [
                51164,
                400,
                570,
                264,
                3687,
                295,
                264,
                5658,
                300,
                321,
                764,
                307,
                257,
                21039,
                278,
                21510,
                11,
                51514
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2133333384990692,
            "compression_ratio": 1.528205128205128,
            "no_speech_prob": 0.008697509765625,
            "words": [
                {
                    "start": 1390.88,
                    "end": 1391.38,
                    "word": " And",
                    "probability": 0.80517578125
                },
                {
                    "start": 1391.38,
                    "end": 1391.88,
                    "word": " because",
                    "probability": 0.7890625
                },
                {
                    "start": 1391.88,
                    "end": 1392.38,
                    "word": " the",
                    "probability": 0.80859375
                },
                {
                    "start": 1392.38,
                    "end": 1392.7,
                    "word": " nature",
                    "probability": 0.77978515625
                },
                {
                    "start": 1392.7,
                    "end": 1393.04,
                    "word": " of",
                    "probability": 0.9013671875
                },
                {
                    "start": 1393.04,
                    "end": 1393.22,
                    "word": " the",
                    "probability": 0.80810546875
                },
                {
                    "start": 1393.22,
                    "end": 1393.66,
                    "word": " noise",
                    "probability": 0.900390625
                },
                {
                    "start": 1393.66,
                    "end": 1393.88,
                    "word": " that",
                    "probability": 0.880859375
                },
                {
                    "start": 1393.88,
                    "end": 1394.02,
                    "word": " we",
                    "probability": 0.9072265625
                },
                {
                    "start": 1394.02,
                    "end": 1394.48,
                    "word": " use",
                    "probability": 0.8193359375
                },
                {
                    "start": 1394.48,
                    "end": 1395.28,
                    "word": " is",
                    "probability": 0.90283203125
                },
                {
                    "start": 1395.46,
                    "end": 1395.82,
                    "word": " a decaying",
                    "probability": 0.830078125
                },
                {
                    "start": 1395.82,
                    "end": 1396.32,
                    "word": " exponential",
                    "probability": 0.9091796875
                },
                {
                    "start": 1396.32,
                    "end": 1398.34,
                    "word": ",",
                    "probability": 0.9697265625
                }
            ]
        },
        {
            "id": 205,
            "seek": 142397,
            "start": 1398.34,
            "end": 1402.34,
            "text": " the delta that is sort of induced by some of the math,",
            "tokens": [
                50364,
                264,
                8289,
                300,
                307,
                1333,
                295,
                33991,
                538,
                512,
                295,
                264,
                5221,
                11,
                50564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1939752322059494,
            "compression_ratio": 1.6888888888888889,
            "no_speech_prob": 0.034088134765625,
            "words": [
                {
                    "start": 1398.34,
                    "end": 1398.64,
                    "word": " the",
                    "probability": 0.2235107421875
                },
                {
                    "start": 1398.64,
                    "end": 1399.74,
                    "word": " delta",
                    "probability": 0.60205078125
                },
                {
                    "start": 1399.74,
                    "end": 1400.22,
                    "word": " that",
                    "probability": 0.8662109375
                },
                {
                    "start": 1400.22,
                    "end": 1400.56,
                    "word": " is",
                    "probability": 0.80322265625
                },
                {
                    "start": 1400.56,
                    "end": 1400.72,
                    "word": " sort",
                    "probability": 0.611328125
                },
                {
                    "start": 1400.72,
                    "end": 1400.82,
                    "word": " of",
                    "probability": 0.90869140625
                },
                {
                    "start": 1400.82,
                    "end": 1401.12,
                    "word": " induced",
                    "probability": 0.61767578125
                },
                {
                    "start": 1401.12,
                    "end": 1401.32,
                    "word": " by",
                    "probability": 0.90234375
                },
                {
                    "start": 1401.32,
                    "end": 1401.52,
                    "word": " some",
                    "probability": 0.81298828125
                },
                {
                    "start": 1401.52,
                    "end": 1401.64,
                    "word": " of",
                    "probability": 0.900390625
                },
                {
                    "start": 1401.64,
                    "end": 1401.76,
                    "word": " the",
                    "probability": 0.8134765625
                },
                {
                    "start": 1401.76,
                    "end": 1402.12,
                    "word": " math",
                    "probability": 0.9189453125
                },
                {
                    "start": 1402.12,
                    "end": 1402.34,
                    "word": ",",
                    "probability": 0.1741943359375
                }
            ]
        },
        {
            "id": 206,
            "seek": 142397,
            "start": 1403.42,
            "end": 1411.12,
            "text": " actually can be sort of smaller than is machine representable in many cases, as we'll get into in a few minutes.",
            "tokens": [
                50614,
                767,
                393,
                312,
                1333,
                295,
                4356,
                813,
                307,
                3479,
                2906,
                712,
                294,
                867,
                3331,
                11,
                382,
                321,
                603,
                483,
                666,
                294,
                257,
                1326,
                2077,
                13,
                50964
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1939752322059494,
            "compression_ratio": 1.6888888888888889,
            "no_speech_prob": 0.034088134765625,
            "words": [
                {
                    "start": 1403.42,
                    "end": 1403.9,
                    "word": " actually",
                    "probability": 0.73876953125
                },
                {
                    "start": 1403.9,
                    "end": 1404.38,
                    "word": " can",
                    "probability": 0.859375
                },
                {
                    "start": 1404.38,
                    "end": 1404.74,
                    "word": " be",
                    "probability": 0.904296875
                },
                {
                    "start": 1404.74,
                    "end": 1405.7,
                    "word": " sort",
                    "probability": 0.60400390625
                },
                {
                    "start": 1405.7,
                    "end": 1405.9,
                    "word": " of",
                    "probability": 0.90380859375
                },
                {
                    "start": 1405.9,
                    "end": 1406.68,
                    "word": " smaller",
                    "probability": 0.75732421875
                },
                {
                    "start": 1406.68,
                    "end": 1406.96,
                    "word": " than",
                    "probability": 0.873046875
                },
                {
                    "start": 1406.96,
                    "end": 1407.16,
                    "word": " is",
                    "probability": 0.83935546875
                },
                {
                    "start": 1407.16,
                    "end": 1407.48,
                    "word": " machine",
                    "probability": 0.73193359375
                },
                {
                    "start": 1407.48,
                    "end": 1408.18,
                    "word": " representable",
                    "probability": 0.82763671875
                },
                {
                    "start": 1408.18,
                    "end": 1408.32,
                    "word": " in",
                    "probability": 0.83447265625
                },
                {
                    "start": 1408.32,
                    "end": 1408.52,
                    "word": " many",
                    "probability": 0.81396484375
                },
                {
                    "start": 1408.52,
                    "end": 1409.0,
                    "word": " cases",
                    "probability": 0.81689453125
                },
                {
                    "start": 1409.0,
                    "end": 1409.24,
                    "word": ",",
                    "probability": 0.77490234375
                },
                {
                    "start": 1409.24,
                    "end": 1409.48,
                    "word": " as",
                    "probability": 0.90185546875
                },
                {
                    "start": 1409.48,
                    "end": 1409.6,
                    "word": " we'll",
                    "probability": 0.920166015625
                },
                {
                    "start": 1409.6,
                    "end": 1409.76,
                    "word": " get",
                    "probability": 0.900390625
                },
                {
                    "start": 1409.76,
                    "end": 1410.0,
                    "word": " into",
                    "probability": 0.7294921875
                },
                {
                    "start": 1410.0,
                    "end": 1410.12,
                    "word": " in",
                    "probability": 0.88427734375
                },
                {
                    "start": 1410.22,
                    "end": 1410.32,
                    "word": " a few",
                    "probability": 0.80615234375
                },
                {
                    "start": 1410.32,
                    "end": 1410.64,
                    "word": " minutes",
                    "probability": 0.88671875
                },
                {
                    "start": 1410.64,
                    "end": 1411.12,
                    "word": ".",
                    "probability": 0.9794921875
                }
            ]
        },
        {
            "id": 207,
            "seek": 142397,
            "start": 1411.22,
            "end": 1414.02,
            "text": " So oftentimes, you don't actually have to worry about a delta,",
            "tokens": [
                51014,
                407,
                18349,
                11,
                291,
                500,
                380,
                767,
                362,
                281,
                3292,
                466,
                257,
                8289,
                11,
                51114
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1939752322059494,
            "compression_ratio": 1.6888888888888889,
            "no_speech_prob": 0.034088134765625,
            "words": [
                {
                    "start": 1411.22,
                    "end": 1411.52,
                    "word": " So",
                    "probability": 0.88916015625
                },
                {
                    "start": 1411.52,
                    "end": 1411.92,
                    "word": " oftentimes",
                    "probability": 0.6669921875
                },
                {
                    "start": 1411.92,
                    "end": 1412.1,
                    "word": ",",
                    "probability": 0.387939453125
                },
                {
                    "start": 1412.1,
                    "end": 1412.3,
                    "word": " you",
                    "probability": 0.89697265625
                },
                {
                    "start": 1412.3,
                    "end": 1412.5,
                    "word": " don't",
                    "probability": 0.947509765625
                },
                {
                    "start": 1412.5,
                    "end": 1412.78,
                    "word": " actually",
                    "probability": 0.79541015625
                },
                {
                    "start": 1412.78,
                    "end": 1412.98,
                    "word": " have",
                    "probability": 0.890625
                },
                {
                    "start": 1412.98,
                    "end": 1413.1,
                    "word": " to",
                    "probability": 0.9072265625
                },
                {
                    "start": 1413.1,
                    "end": 1413.24,
                    "word": " worry",
                    "probability": 0.89794921875
                },
                {
                    "start": 1413.24,
                    "end": 1413.44,
                    "word": " about",
                    "probability": 0.8134765625
                },
                {
                    "start": 1413.56,
                    "end": 1413.78,
                    "word": " a delta",
                    "probability": 0.8388671875
                },
                {
                    "start": 1413.78,
                    "end": 1414.02,
                    "word": ",",
                    "probability": 0.77880859375
                }
            ]
        },
        {
            "id": 208,
            "seek": 142397,
            "start": 1414.02,
            "end": 1417.82,
            "text": " but it's sort of important enough that you should be able to reason about it.",
            "tokens": [
                51164,
                457,
                309,
                311,
                1333,
                295,
                1021,
                1547,
                300,
                291,
                820,
                312,
                1075,
                281,
                1778,
                466,
                309,
                13,
                51314
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1939752322059494,
            "compression_ratio": 1.6888888888888889,
            "no_speech_prob": 0.034088134765625,
            "words": [
                {
                    "start": 1414.02,
                    "end": 1414.08,
                    "word": " but",
                    "probability": 0.904296875
                },
                {
                    "start": 1414.08,
                    "end": 1414.46,
                    "word": " it's",
                    "probability": 0.9443359375
                },
                {
                    "start": 1414.46,
                    "end": 1414.58,
                    "word": " sort",
                    "probability": 0.7802734375
                },
                {
                    "start": 1414.58,
                    "end": 1414.66,
                    "word": " of",
                    "probability": 0.90625
                },
                {
                    "start": 1414.66,
                    "end": 1414.98,
                    "word": " important",
                    "probability": 0.76953125
                },
                {
                    "start": 1414.98,
                    "end": 1415.44,
                    "word": " enough",
                    "probability": 0.77978515625
                },
                {
                    "start": 1415.44,
                    "end": 1416.16,
                    "word": " that",
                    "probability": 0.806640625
                },
                {
                    "start": 1416.16,
                    "end": 1416.24,
                    "word": " you",
                    "probability": 0.896484375
                },
                {
                    "start": 1416.24,
                    "end": 1416.38,
                    "word": " should",
                    "probability": 0.9228515625
                },
                {
                    "start": 1416.38,
                    "end": 1416.46,
                    "word": " be",
                    "probability": 0.90966796875
                },
                {
                    "start": 1416.46,
                    "end": 1416.58,
                    "word": " able",
                    "probability": 0.89111328125
                },
                {
                    "start": 1416.58,
                    "end": 1416.7,
                    "word": " to",
                    "probability": 0.904296875
                },
                {
                    "start": 1416.7,
                    "end": 1416.88,
                    "word": " reason",
                    "probability": 0.90771484375
                },
                {
                    "start": 1416.88,
                    "end": 1417.16,
                    "word": " about",
                    "probability": 0.81787109375
                },
                {
                    "start": 1417.16,
                    "end": 1417.34,
                    "word": " it",
                    "probability": 0.90380859375
                },
                {
                    "start": 1417.34,
                    "end": 1417.82,
                    "word": ".",
                    "probability": 0.99755859375
                }
            ]
        },
        {
            "id": 209,
            "seek": 142397,
            "start": 1418.42,
            "end": 1422.78,
            "text": " So just returning earlier, we talked about the privacy model,",
            "tokens": [
                51364,
                407,
                445,
                12678,
                3071,
                11,
                321,
                2825,
                466,
                264,
                11427,
                2316,
                11,
                51564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1939752322059494,
            "compression_ratio": 1.6888888888888889,
            "no_speech_prob": 0.034088134765625,
            "words": [
                {
                    "start": 1418.42,
                    "end": 1418.9,
                    "word": " So",
                    "probability": 0.88818359375
                },
                {
                    "start": 1418.9,
                    "end": 1419.12,
                    "word": " just",
                    "probability": 0.810546875
                },
                {
                    "start": 1419.12,
                    "end": 1419.44,
                    "word": " returning",
                    "probability": 0.81298828125
                },
                {
                    "start": 1419.44,
                    "end": 1419.82,
                    "word": " earlier",
                    "probability": 0.90869140625
                },
                {
                    "start": 1419.82,
                    "end": 1419.96,
                    "word": ",",
                    "probability": 0.9423828125
                },
                {
                    "start": 1419.96,
                    "end": 1420.04,
                    "word": " we",
                    "probability": 0.90380859375
                },
                {
                    "start": 1420.04,
                    "end": 1420.32,
                    "word": " talked",
                    "probability": 0.74560546875
                },
                {
                    "start": 1420.32,
                    "end": 1420.62,
                    "word": " about",
                    "probability": 0.8203125
                },
                {
                    "start": 1420.62,
                    "end": 1421.0,
                    "word": " the",
                    "probability": 0.814453125
                },
                {
                    "start": 1421.0,
                    "end": 1422.02,
                    "word": " privacy",
                    "probability": 0.82666015625
                },
                {
                    "start": 1422.02,
                    "end": 1422.44,
                    "word": " model",
                    "probability": 0.8154296875
                },
                {
                    "start": 1422.44,
                    "end": 1422.78,
                    "word": ",",
                    "probability": 0.7919921875
                }
            ]
        },
        {
            "id": 210,
            "seek": 142397,
            "start": 1422.78,
            "end": 1427.58,
            "text": " per-user transformation that's very generous, more restricted cross-user aggregation.",
            "tokens": [
                51614,
                680,
                12,
                18088,
                9887,
                300,
                311,
                588,
                14537,
                11,
                544,
                20608,
                3278,
                12,
                18088,
                16743,
                399,
                13,
                51814
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1939752322059494,
            "compression_ratio": 1.6888888888888889,
            "no_speech_prob": 0.034088134765625,
            "words": [
                {
                    "start": 1422.78,
                    "end": 1423.18,
                    "word": " per",
                    "probability": 0.50146484375
                },
                {
                    "start": 1423.18,
                    "end": 1423.42,
                    "word": "-user",
                    "probability": 0.6029052734375
                },
                {
                    "start": 1423.42,
                    "end": 1424.06,
                    "word": " transformation",
                    "probability": 0.7861328125
                },
                {
                    "start": 1424.06,
                    "end": 1424.54,
                    "word": " that's",
                    "probability": 0.776611328125
                },
                {
                    "start": 1424.54,
                    "end": 1424.76,
                    "word": " very",
                    "probability": 0.7431640625
                },
                {
                    "start": 1424.76,
                    "end": 1425.2,
                    "word": " generous",
                    "probability": 0.8525390625
                },
                {
                    "start": 1425.2,
                    "end": 1425.56,
                    "word": ",",
                    "probability": 0.92919921875
                },
                {
                    "start": 1425.56,
                    "end": 1425.82,
                    "word": " more",
                    "probability": 0.8720703125
                },
                {
                    "start": 1425.82,
                    "end": 1426.24,
                    "word": " restricted",
                    "probability": 0.61328125
                },
                {
                    "start": 1426.24,
                    "end": 1426.58,
                    "word": " cross",
                    "probability": 0.8193359375
                },
                {
                    "start": 1426.58,
                    "end": 1426.8,
                    "word": "-user",
                    "probability": 0.861328125
                },
                {
                    "start": 1426.8,
                    "end": 1427.48,
                    "word": " aggregation",
                    "probability": 0.872314453125
                },
                {
                    "start": 1427.48,
                    "end": 1427.58,
                    "word": ".",
                    "probability": 0.96875
                }
            ]
        },
        {
            "id": 211,
            "seek": 145075,
            "start": 1428.34,
            "end": 1429.54,
            "text": " And that's why we apply to SQL.",
            "tokens": [
                50364,
                400,
                300,
                311,
                983,
                321,
                3079,
                281,
                19200,
                13,
                50414
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2832236854653609,
            "compression_ratio": 1.55078125,
            "no_speech_prob": 0.10369873046875,
            "words": [
                {
                    "start": 1428.34,
                    "end": 1428.42,
                    "word": " And",
                    "probability": 0.0107574462890625
                },
                {
                    "start": 1428.42,
                    "end": 1428.42,
                    "word": " that's",
                    "probability": 0.265289306640625
                },
                {
                    "start": 1428.42,
                    "end": 1428.42,
                    "word": " why",
                    "probability": 0.0745849609375
                },
                {
                    "start": 1428.42,
                    "end": 1428.42,
                    "word": " we",
                    "probability": 0.439697265625
                },
                {
                    "start": 1428.42,
                    "end": 1428.42,
                    "word": " apply",
                    "probability": 0.0216522216796875
                },
                {
                    "start": 1428.42,
                    "end": 1428.6,
                    "word": " to",
                    "probability": 0.556640625
                },
                {
                    "start": 1428.6,
                    "end": 1428.9,
                    "word": " SQL",
                    "probability": 0.87548828125
                },
                {
                    "start": 1428.9,
                    "end": 1429.54,
                    "word": ".",
                    "probability": 0.9697265625
                }
            ]
        },
        {
            "id": 212,
            "seek": 145075,
            "start": 1429.96,
            "end": 1437.72,
            "text": " Let's take kind of an example of a typical privacy-sensitive query that seems pretty innocuous.",
            "tokens": [
                50464,
                961,
                311,
                747,
                733,
                295,
                364,
                1365,
                295,
                257,
                7476,
                11427,
                12,
                82,
                34465,
                14581,
                300,
                2544,
                1238,
                10843,
                12549,
                13,
                50814
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2832236854653609,
            "compression_ratio": 1.55078125,
            "no_speech_prob": 0.10369873046875,
            "words": [
                {
                    "start": 1429.96,
                    "end": 1430.42,
                    "word": " Let's",
                    "probability": 0.923095703125
                },
                {
                    "start": 1430.42,
                    "end": 1430.62,
                    "word": " take",
                    "probability": 0.79833984375
                },
                {
                    "start": 1430.62,
                    "end": 1430.78,
                    "word": " kind",
                    "probability": 0.5546875
                },
                {
                    "start": 1430.78,
                    "end": 1430.84,
                    "word": " of",
                    "probability": 0.90869140625
                },
                {
                    "start": 1430.84,
                    "end": 1430.94,
                    "word": " an",
                    "probability": 0.89990234375
                },
                {
                    "start": 1430.94,
                    "end": 1431.32,
                    "word": " example",
                    "probability": 0.9296875
                },
                {
                    "start": 1431.32,
                    "end": 1432.14,
                    "word": " of",
                    "probability": 0.9033203125
                },
                {
                    "start": 1432.88,
                    "end": 1433.7,
                    "word": " a typical",
                    "probability": 0.53173828125
                },
                {
                    "start": 1433.7,
                    "end": 1435.18,
                    "word": " privacy",
                    "probability": 0.84814453125
                },
                {
                    "start": 1435.18,
                    "end": 1435.7,
                    "word": "-sensitive",
                    "probability": 0.7036946614583334
                },
                {
                    "start": 1435.7,
                    "end": 1436.12,
                    "word": " query",
                    "probability": 0.8984375
                },
                {
                    "start": 1436.12,
                    "end": 1436.32,
                    "word": " that",
                    "probability": 0.857421875
                },
                {
                    "start": 1436.32,
                    "end": 1436.56,
                    "word": " seems",
                    "probability": 0.7763671875
                },
                {
                    "start": 1436.56,
                    "end": 1436.84,
                    "word": " pretty",
                    "probability": 0.90625
                },
                {
                    "start": 1436.84,
                    "end": 1437.42,
                    "word": " innocuous",
                    "probability": 0.950927734375
                },
                {
                    "start": 1437.42,
                    "end": 1437.72,
                    "word": ".",
                    "probability": 0.60791015625
                }
            ]
        },
        {
            "id": 213,
            "seek": 145075,
            "start": 1437.72,
            "end": 1443.58,
            "text": " We're counting up the number of visits to a website based on browser agent.",
            "tokens": [
                50864,
                492,
                434,
                13251,
                493,
                264,
                1230,
                295,
                17753,
                281,
                257,
                3144,
                2361,
                322,
                11185,
                9461,
                13,
                51114
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2832236854653609,
            "compression_ratio": 1.55078125,
            "no_speech_prob": 0.10369873046875,
            "words": [
                {
                    "start": 1437.72,
                    "end": 1438.22,
                    "word": " We're",
                    "probability": 0.8671875
                },
                {
                    "start": 1438.22,
                    "end": 1438.48,
                    "word": " counting",
                    "probability": 0.74365234375
                },
                {
                    "start": 1438.48,
                    "end": 1438.74,
                    "word": " up",
                    "probability": 0.8984375
                },
                {
                    "start": 1438.74,
                    "end": 1438.88,
                    "word": " the",
                    "probability": 0.8095703125
                },
                {
                    "start": 1438.88,
                    "end": 1439.22,
                    "word": " number",
                    "probability": 0.87841796875
                },
                {
                    "start": 1439.22,
                    "end": 1439.9,
                    "word": " of",
                    "probability": 0.90185546875
                },
                {
                    "start": 1439.9,
                    "end": 1440.18,
                    "word": " visits",
                    "probability": 0.92431640625
                },
                {
                    "start": 1440.18,
                    "end": 1440.56,
                    "word": " to",
                    "probability": 0.8974609375
                },
                {
                    "start": 1440.84,
                    "end": 1441.22,
                    "word": " a website",
                    "probability": 0.80419921875
                },
                {
                    "start": 1441.22,
                    "end": 1442.06,
                    "word": " based",
                    "probability": 0.7958984375
                },
                {
                    "start": 1442.06,
                    "end": 1442.3,
                    "word": " on",
                    "probability": 0.90380859375
                },
                {
                    "start": 1442.3,
                    "end": 1442.62,
                    "word": " browser",
                    "probability": 0.53955078125
                },
                {
                    "start": 1442.62,
                    "end": 1442.94,
                    "word": " agent",
                    "probability": 0.62109375
                },
                {
                    "start": 1442.94,
                    "end": 1443.58,
                    "word": ".",
                    "probability": 0.98388671875
                }
            ]
        },
        {
            "id": 214,
            "seek": 145075,
            "start": 1443.8,
            "end": 1449.62,
            "text": " So a quick look through the logs reveals that a single user used a particular agent.",
            "tokens": [
                51164,
                407,
                257,
                1702,
                574,
                807,
                264,
                20820,
                20893,
                300,
                257,
                2167,
                4195,
                1143,
                257,
                1729,
                9461,
                13,
                51414
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2832236854653609,
            "compression_ratio": 1.55078125,
            "no_speech_prob": 0.10369873046875,
            "words": [
                {
                    "start": 1443.8,
                    "end": 1444.04,
                    "word": " So",
                    "probability": 0.83935546875
                },
                {
                    "start": 1444.28,
                    "end": 1444.46,
                    "word": " a quick",
                    "probability": 0.91796875
                },
                {
                    "start": 1444.46,
                    "end": 1444.8,
                    "word": " look",
                    "probability": 0.8974609375
                },
                {
                    "start": 1444.8,
                    "end": 1445.14,
                    "word": " through",
                    "probability": 0.7919921875
                },
                {
                    "start": 1445.14,
                    "end": 1445.52,
                    "word": " the",
                    "probability": 0.81103515625
                },
                {
                    "start": 1445.52,
                    "end": 1445.88,
                    "word": " logs",
                    "probability": 0.7197265625
                },
                {
                    "start": 1445.88,
                    "end": 1446.28,
                    "word": " reveals",
                    "probability": 0.91552734375
                },
                {
                    "start": 1446.28,
                    "end": 1446.72,
                    "word": " that",
                    "probability": 0.86376953125
                },
                {
                    "start": 1446.94,
                    "end": 1447.2,
                    "word": " a single",
                    "probability": 0.8828125
                },
                {
                    "start": 1447.2,
                    "end": 1447.6,
                    "word": " user",
                    "probability": 0.8046875
                },
                {
                    "start": 1447.6,
                    "end": 1448.0,
                    "word": " used",
                    "probability": 0.7939453125
                },
                {
                    "start": 1448.22,
                    "end": 1448.58,
                    "word": " a particular",
                    "probability": 0.8076171875
                },
                {
                    "start": 1448.58,
                    "end": 1449.0,
                    "word": " agent",
                    "probability": 0.81396484375
                },
                {
                    "start": 1449.0,
                    "end": 1449.62,
                    "word": ".",
                    "probability": 0.61376953125
                }
            ]
        },
        {
            "id": 215,
            "seek": 145075,
            "start": 1449.9,
            "end": 1455.12,
            "text": " And already, we've kind of re-identified an individual without doing anything more than counting statistics.",
            "tokens": [
                51464,
                400,
                1217,
                11,
                321,
                600,
                733,
                295,
                319,
                12,
                40613,
                364,
                2609,
                1553,
                884,
                1340,
                544,
                813,
                13251,
                12523,
                13,
                51714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2832236854653609,
            "compression_ratio": 1.55078125,
            "no_speech_prob": 0.10369873046875,
            "words": [
                {
                    "start": 1449.9,
                    "end": 1450.54,
                    "word": " And",
                    "probability": 0.892578125
                },
                {
                    "start": 1450.54,
                    "end": 1450.86,
                    "word": " already",
                    "probability": 0.8388671875
                },
                {
                    "start": 1450.86,
                    "end": 1450.96,
                    "word": ",",
                    "probability": 0.332275390625
                },
                {
                    "start": 1450.96,
                    "end": 1451.16,
                    "word": " we've",
                    "probability": 0.925537109375
                },
                {
                    "start": 1451.16,
                    "end": 1451.3,
                    "word": " kind",
                    "probability": 0.76025390625
                },
                {
                    "start": 1451.3,
                    "end": 1451.4,
                    "word": " of",
                    "probability": 0.90087890625
                },
                {
                    "start": 1451.4,
                    "end": 1451.58,
                    "word": " re",
                    "probability": 0.9375
                },
                {
                    "start": 1451.58,
                    "end": 1451.98,
                    "word": "-identified",
                    "probability": 0.708984375
                },
                {
                    "start": 1451.98,
                    "end": 1452.3,
                    "word": " an",
                    "probability": 0.90087890625
                },
                {
                    "start": 1452.3,
                    "end": 1452.66,
                    "word": " individual",
                    "probability": 0.74658203125
                },
                {
                    "start": 1452.66,
                    "end": 1453.0,
                    "word": " without",
                    "probability": 0.8076171875
                },
                {
                    "start": 1453.0,
                    "end": 1453.26,
                    "word": " doing",
                    "probability": 0.91552734375
                },
                {
                    "start": 1453.26,
                    "end": 1453.52,
                    "word": " anything",
                    "probability": 0.8349609375
                },
                {
                    "start": 1453.52,
                    "end": 1453.8,
                    "word": " more",
                    "probability": 0.8759765625
                },
                {
                    "start": 1453.8,
                    "end": 1453.96,
                    "word": " than",
                    "probability": 0.8828125
                },
                {
                    "start": 1453.96,
                    "end": 1454.28,
                    "word": " counting",
                    "probability": 0.73779296875
                },
                {
                    "start": 1454.28,
                    "end": 1454.84,
                    "word": " statistics",
                    "probability": 0.90087890625
                },
                {
                    "start": 1454.84,
                    "end": 1455.12,
                    "word": ".",
                    "probability": 0.99169921875
                }
            ]
        },
        {
            "id": 216,
            "seek": 147951,
            "start": 1455.12,
            "end": 1460.0,
            "text": " So we can see the problem with kind of a non-differentially private environment already.",
            "tokens": [
                50364,
                407,
                321,
                393,
                536,
                264,
                1154,
                365,
                733,
                295,
                257,
                2107,
                12,
                67,
                12612,
                3137,
                4551,
                2823,
                1217,
                13,
                50564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1785511382601478,
            "compression_ratio": 1.7244094488188977,
            "no_speech_prob": 0.003765106201171875,
            "words": [
                {
                    "start": 1455.12,
                    "end": 1455.24,
                    "word": " So",
                    "probability": 0.76416015625
                },
                {
                    "start": 1455.24,
                    "end": 1455.36,
                    "word": " we",
                    "probability": 0.8046875
                },
                {
                    "start": 1455.36,
                    "end": 1455.46,
                    "word": " can",
                    "probability": 0.90087890625
                },
                {
                    "start": 1455.46,
                    "end": 1455.56,
                    "word": " see",
                    "probability": 0.81787109375
                },
                {
                    "start": 1455.56,
                    "end": 1455.72,
                    "word": " the",
                    "probability": 0.81640625
                },
                {
                    "start": 1455.72,
                    "end": 1455.98,
                    "word": " problem",
                    "probability": 0.7685546875
                },
                {
                    "start": 1455.98,
                    "end": 1456.12,
                    "word": " with",
                    "probability": 0.7900390625
                },
                {
                    "start": 1456.12,
                    "end": 1456.26,
                    "word": " kind",
                    "probability": 0.5830078125
                },
                {
                    "start": 1456.26,
                    "end": 1456.34,
                    "word": " of",
                    "probability": 0.9033203125
                },
                {
                    "start": 1456.54,
                    "end": 1456.94,
                    "word": " a non",
                    "probability": 0.8896484375
                },
                {
                    "start": 1456.94,
                    "end": 1457.32,
                    "word": "-differentially",
                    "probability": 0.808349609375
                },
                {
                    "start": 1457.32,
                    "end": 1457.74,
                    "word": " private",
                    "probability": 0.7900390625
                },
                {
                    "start": 1457.74,
                    "end": 1458.58,
                    "word": " environment",
                    "probability": 0.724609375
                },
                {
                    "start": 1458.58,
                    "end": 1459.52,
                    "word": " already",
                    "probability": 0.7890625
                },
                {
                    "start": 1459.52,
                    "end": 1460.0,
                    "word": ".",
                    "probability": 0.9833984375
                }
            ]
        },
        {
            "id": 217,
            "seek": 147951,
            "start": 1460.44,
            "end": 1466.78,
            "text": " So classically, sort of naively, one thing that one does in differential privacy",
            "tokens": [
                50614,
                407,
                1508,
                984,
                11,
                1333,
                295,
                1667,
                3413,
                11,
                472,
                551,
                300,
                472,
                775,
                294,
                15756,
                11427,
                50914
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1785511382601478,
            "compression_ratio": 1.7244094488188977,
            "no_speech_prob": 0.003765106201171875,
            "words": [
                {
                    "start": 1460.44,
                    "end": 1460.7,
                    "word": " So",
                    "probability": 0.8349609375
                },
                {
                    "start": 1460.7,
                    "end": 1463.04,
                    "word": " classically",
                    "probability": 0.7354736328125
                },
                {
                    "start": 1463.04,
                    "end": 1463.18,
                    "word": ",",
                    "probability": 0.81298828125
                },
                {
                    "start": 1463.18,
                    "end": 1463.3,
                    "word": " sort",
                    "probability": 0.7587890625
                },
                {
                    "start": 1463.3,
                    "end": 1463.38,
                    "word": " of",
                    "probability": 0.8984375
                },
                {
                    "start": 1463.38,
                    "end": 1463.76,
                    "word": " naively",
                    "probability": 0.829833984375
                },
                {
                    "start": 1463.76,
                    "end": 1464.38,
                    "word": ",",
                    "probability": 0.916015625
                },
                {
                    "start": 1464.38,
                    "end": 1465.0,
                    "word": " one",
                    "probability": 0.81689453125
                },
                {
                    "start": 1465.0,
                    "end": 1465.18,
                    "word": " thing",
                    "probability": 0.8115234375
                },
                {
                    "start": 1465.18,
                    "end": 1465.36,
                    "word": " that",
                    "probability": 0.88037109375
                },
                {
                    "start": 1465.36,
                    "end": 1465.56,
                    "word": " one",
                    "probability": 0.8125
                },
                {
                    "start": 1465.56,
                    "end": 1465.94,
                    "word": " does",
                    "probability": 0.92626953125
                },
                {
                    "start": 1465.94,
                    "end": 1466.12,
                    "word": " in",
                    "probability": 0.9111328125
                },
                {
                    "start": 1466.12,
                    "end": 1466.42,
                    "word": " differential",
                    "probability": 0.732421875
                },
                {
                    "start": 1466.42,
                    "end": 1466.78,
                    "word": " privacy",
                    "probability": 0.84130859375
                }
            ]
        },
        {
            "id": 218,
            "seek": 147951,
            "start": 1466.78,
            "end": 1472.0,
            "text": " is they add some kind of noise that satisfies differential privacy to counts.",
            "tokens": [
                50964,
                307,
                436,
                909,
                512,
                733,
                295,
                5658,
                300,
                44271,
                15756,
                11427,
                281,
                14893,
                13,
                51164
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1785511382601478,
            "compression_ratio": 1.7244094488188977,
            "no_speech_prob": 0.003765106201171875,
            "words": [
                {
                    "start": 1466.78,
                    "end": 1467.04,
                    "word": " is",
                    "probability": 0.861328125
                },
                {
                    "start": 1467.04,
                    "end": 1467.16,
                    "word": " they",
                    "probability": 0.72802734375
                },
                {
                    "start": 1467.16,
                    "end": 1467.46,
                    "word": " add",
                    "probability": 0.802734375
                },
                {
                    "start": 1467.46,
                    "end": 1467.8,
                    "word": " some",
                    "probability": 0.81396484375
                },
                {
                    "start": 1467.8,
                    "end": 1468.18,
                    "word": " kind",
                    "probability": 0.80517578125
                },
                {
                    "start": 1468.18,
                    "end": 1468.32,
                    "word": " of",
                    "probability": 0.904296875
                },
                {
                    "start": 1468.32,
                    "end": 1468.66,
                    "word": " noise",
                    "probability": 0.8974609375
                },
                {
                    "start": 1468.66,
                    "end": 1469.0,
                    "word": " that",
                    "probability": 0.8828125
                },
                {
                    "start": 1469.0,
                    "end": 1469.54,
                    "word": " satisfies",
                    "probability": 0.796875
                },
                {
                    "start": 1469.54,
                    "end": 1469.98,
                    "word": " differential",
                    "probability": 0.76953125
                },
                {
                    "start": 1469.98,
                    "end": 1470.56,
                    "word": " privacy",
                    "probability": 0.8447265625
                },
                {
                    "start": 1470.56,
                    "end": 1471.32,
                    "word": " to",
                    "probability": 0.86279296875
                },
                {
                    "start": 1471.32,
                    "end": 1471.68,
                    "word": " counts",
                    "probability": 0.78759765625
                },
                {
                    "start": 1471.68,
                    "end": 1472.0,
                    "word": ".",
                    "probability": 0.97216796875
                }
            ]
        },
        {
            "id": 219,
            "seek": 147951,
            "start": 1472.0,
            "end": 1475.02,
            "text": " And so in this case, we go through all the browser agents,",
            "tokens": [
                51214,
                400,
                370,
                294,
                341,
                1389,
                11,
                321,
                352,
                807,
                439,
                264,
                11185,
                12554,
                11,
                51314
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1785511382601478,
            "compression_ratio": 1.7244094488188977,
            "no_speech_prob": 0.003765106201171875,
            "words": [
                {
                    "start": 1472.0,
                    "end": 1472.12,
                    "word": " And",
                    "probability": 0.81005859375
                },
                {
                    "start": 1472.12,
                    "end": 1472.42,
                    "word": " so",
                    "probability": 0.90234375
                },
                {
                    "start": 1472.42,
                    "end": 1472.66,
                    "word": " in",
                    "probability": 0.80224609375
                },
                {
                    "start": 1472.66,
                    "end": 1472.84,
                    "word": " this",
                    "probability": 0.88232421875
                },
                {
                    "start": 1472.84,
                    "end": 1473.12,
                    "word": " case",
                    "probability": 0.814453125
                },
                {
                    "start": 1473.12,
                    "end": 1473.14,
                    "word": ",",
                    "probability": 0.92724609375
                },
                {
                    "start": 1473.14,
                    "end": 1473.36,
                    "word": " we",
                    "probability": 0.90380859375
                },
                {
                    "start": 1473.36,
                    "end": 1473.6,
                    "word": " go",
                    "probability": 0.90087890625
                },
                {
                    "start": 1473.6,
                    "end": 1473.76,
                    "word": " through",
                    "probability": 0.79833984375
                },
                {
                    "start": 1473.76,
                    "end": 1473.9,
                    "word": " all",
                    "probability": 0.90087890625
                },
                {
                    "start": 1473.9,
                    "end": 1474.02,
                    "word": " the",
                    "probability": 0.8154296875
                },
                {
                    "start": 1474.02,
                    "end": 1474.22,
                    "word": " browser",
                    "probability": 0.7490234375
                },
                {
                    "start": 1474.22,
                    "end": 1474.68,
                    "word": " agents",
                    "probability": 0.77392578125
                },
                {
                    "start": 1474.68,
                    "end": 1475.02,
                    "word": ",",
                    "probability": 0.5986328125
                }
            ]
        },
        {
            "id": 220,
            "seek": 147951,
            "start": 1475.02,
            "end": 1480.82,
            "text": " we adjust the counts, plus or minus, a small amount of noise dependent on epsilon.",
            "tokens": [
                51364,
                321,
                4369,
                264,
                14893,
                11,
                1804,
                420,
                3175,
                11,
                257,
                1359,
                2372,
                295,
                5658,
                12334,
                322,
                17889,
                13,
                51614
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1785511382601478,
            "compression_ratio": 1.7244094488188977,
            "no_speech_prob": 0.003765106201171875,
            "words": [
                {
                    "start": 1475.02,
                    "end": 1475.22,
                    "word": " we",
                    "probability": 0.90234375
                },
                {
                    "start": 1475.22,
                    "end": 1475.48,
                    "word": " adjust",
                    "probability": 0.8466796875
                },
                {
                    "start": 1475.48,
                    "end": 1475.66,
                    "word": " the",
                    "probability": 0.81103515625
                },
                {
                    "start": 1475.66,
                    "end": 1476.02,
                    "word": " counts",
                    "probability": 0.71337890625
                },
                {
                    "start": 1476.02,
                    "end": 1476.14,
                    "word": ",",
                    "probability": 0.489013671875
                },
                {
                    "start": 1476.14,
                    "end": 1476.28,
                    "word": " plus",
                    "probability": 0.88232421875
                },
                {
                    "start": 1476.28,
                    "end": 1476.44,
                    "word": " or",
                    "probability": 0.89599609375
                },
                {
                    "start": 1476.44,
                    "end": 1476.78,
                    "word": " minus",
                    "probability": 0.92724609375
                },
                {
                    "start": 1476.78,
                    "end": 1477.16,
                    "word": ",",
                    "probability": 0.5107421875
                },
                {
                    "start": 1477.4,
                    "end": 1477.68,
                    "word": " a small",
                    "probability": 0.85009765625
                },
                {
                    "start": 1477.68,
                    "end": 1477.88,
                    "word": " amount",
                    "probability": 0.84912109375
                },
                {
                    "start": 1477.88,
                    "end": 1478.04,
                    "word": " of",
                    "probability": 0.904296875
                },
                {
                    "start": 1478.04,
                    "end": 1478.42,
                    "word": " noise",
                    "probability": 0.8779296875
                },
                {
                    "start": 1478.42,
                    "end": 1479.82,
                    "word": " dependent",
                    "probability": 0.611328125
                },
                {
                    "start": 1479.82,
                    "end": 1480.04,
                    "word": " on",
                    "probability": 0.904296875
                },
                {
                    "start": 1480.04,
                    "end": 1480.34,
                    "word": " epsilon",
                    "probability": 0.5966796875
                },
                {
                    "start": 1480.34,
                    "end": 1480.82,
                    "word": ".",
                    "probability": 0.96826171875
                }
            ]
        },
        {
            "id": 221,
            "seek": 147951,
            "start": 1481.54,
            "end": 1483.88,
            "text": " And in general, you might think, oh, we're good.",
            "tokens": [
                51664,
                400,
                294,
                2674,
                11,
                291,
                1062,
                519,
                11,
                1954,
                11,
                321,
                434,
                665,
                13,
                51764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1785511382601478,
            "compression_ratio": 1.7244094488188977,
            "no_speech_prob": 0.003765106201171875,
            "words": [
                {
                    "start": 1481.54,
                    "end": 1482.02,
                    "word": " And",
                    "probability": 0.892578125
                },
                {
                    "start": 1482.02,
                    "end": 1482.18,
                    "word": " in",
                    "probability": 0.89306640625
                },
                {
                    "start": 1482.18,
                    "end": 1482.44,
                    "word": " general",
                    "probability": 0.802734375
                },
                {
                    "start": 1482.44,
                    "end": 1482.58,
                    "word": ",",
                    "probability": 0.92236328125
                },
                {
                    "start": 1482.58,
                    "end": 1482.78,
                    "word": " you",
                    "probability": 0.890625
                },
                {
                    "start": 1482.78,
                    "end": 1482.94,
                    "word": " might",
                    "probability": 0.81298828125
                },
                {
                    "start": 1482.94,
                    "end": 1483.2,
                    "word": " think",
                    "probability": 0.8271484375
                },
                {
                    "start": 1483.2,
                    "end": 1483.28,
                    "word": ",",
                    "probability": 0.95751953125
                },
                {
                    "start": 1483.28,
                    "end": 1483.38,
                    "word": " oh",
                    "probability": 0.55810546875
                },
                {
                    "start": 1483.38,
                    "end": 1483.4,
                    "word": ",",
                    "probability": 0.943359375
                },
                {
                    "start": 1483.4,
                    "end": 1483.54,
                    "word": " we're",
                    "probability": 0.927978515625
                },
                {
                    "start": 1483.54,
                    "end": 1483.86,
                    "word": " good",
                    "probability": 0.87451171875
                },
                {
                    "start": 1483.86,
                    "end": 1483.88,
                    "word": ".",
                    "probability": 0.402587890625
                }
            ]
        },
        {
            "id": 222,
            "seek": 150869,
            "start": 1483.88,
            "end": 1489.5,
            "text": " Except that a browser agent can be an arbitrary string, right?",
            "tokens": [
                50364,
                16192,
                300,
                257,
                11185,
                9461,
                393,
                312,
                364,
                23211,
                6798,
                11,
                558,
                30,
                50614
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19074876119594764,
            "compression_ratio": 1.6337448559670782,
            "no_speech_prob": 0.0025119781494140625,
            "words": [
                {
                    "start": 1483.88,
                    "end": 1484.6,
                    "word": " Except",
                    "probability": 0.073974609375
                },
                {
                    "start": 1484.6,
                    "end": 1486.56,
                    "word": " that",
                    "probability": 0.69873046875
                },
                {
                    "start": 1487.18,
                    "end": 1487.38,
                    "word": " a browser",
                    "probability": 0.736328125
                },
                {
                    "start": 1487.38,
                    "end": 1487.68,
                    "word": " agent",
                    "probability": 0.81787109375
                },
                {
                    "start": 1487.68,
                    "end": 1487.88,
                    "word": " can",
                    "probability": 0.89208984375
                },
                {
                    "start": 1487.88,
                    "end": 1488.1,
                    "word": " be",
                    "probability": 0.9052734375
                },
                {
                    "start": 1488.1,
                    "end": 1488.3,
                    "word": " an",
                    "probability": 0.892578125
                },
                {
                    "start": 1488.3,
                    "end": 1488.6,
                    "word": " arbitrary",
                    "probability": 0.771484375
                },
                {
                    "start": 1488.6,
                    "end": 1489.04,
                    "word": " string",
                    "probability": 0.78173828125
                },
                {
                    "start": 1489.04,
                    "end": 1489.24,
                    "word": ",",
                    "probability": 0.53857421875
                },
                {
                    "start": 1489.24,
                    "end": 1489.4,
                    "word": " right",
                    "probability": 0.8203125
                },
                {
                    "start": 1489.4,
                    "end": 1489.5,
                    "word": "?",
                    "probability": 0.7294921875
                }
            ]
        },
        {
            "id": 223,
            "seek": 150869,
            "start": 1489.5,
            "end": 1492.0,
            "text": " So this could be a made-up thing.",
            "tokens": [
                50664,
                407,
                341,
                727,
                312,
                257,
                1027,
                12,
                1010,
                551,
                13,
                50714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19074876119594764,
            "compression_ratio": 1.6337448559670782,
            "no_speech_prob": 0.0025119781494140625,
            "words": [
                {
                    "start": 1489.5,
                    "end": 1489.6,
                    "word": " So",
                    "probability": 0.87939453125
                },
                {
                    "start": 1489.6,
                    "end": 1489.78,
                    "word": " this",
                    "probability": 0.7109375
                },
                {
                    "start": 1489.78,
                    "end": 1489.92,
                    "word": " could",
                    "probability": 0.79150390625
                },
                {
                    "start": 1489.92,
                    "end": 1490.3,
                    "word": " be",
                    "probability": 0.89794921875
                },
                {
                    "start": 1490.94,
                    "end": 1491.16,
                    "word": " a made",
                    "probability": 0.83154296875
                },
                {
                    "start": 1491.16,
                    "end": 1491.28,
                    "word": "-up",
                    "probability": 0.76171875
                },
                {
                    "start": 1491.28,
                    "end": 1491.64,
                    "word": " thing",
                    "probability": 0.80712890625
                },
                {
                    "start": 1491.64,
                    "end": 1492.0,
                    "word": ".",
                    "probability": 0.71240234375
                }
            ]
        },
        {
            "id": 224,
            "seek": 150869,
            "start": 1492.0,
            "end": 1495.12,
            "text": " Essentially, from the position of an attacker,",
            "tokens": [
                50764,
                23596,
                11,
                490,
                264,
                2535,
                295,
                364,
                35871,
                11,
                50864
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19074876119594764,
            "compression_ratio": 1.6337448559670782,
            "no_speech_prob": 0.0025119781494140625,
            "words": [
                {
                    "start": 1492.0,
                    "end": 1492.32,
                    "word": " Essentially",
                    "probability": 0.82080078125
                },
                {
                    "start": 1492.32,
                    "end": 1492.58,
                    "word": ",",
                    "probability": 0.60205078125
                },
                {
                    "start": 1492.58,
                    "end": 1492.82,
                    "word": " from",
                    "probability": 0.8125
                },
                {
                    "start": 1492.82,
                    "end": 1493.08,
                    "word": " the",
                    "probability": 0.81298828125
                },
                {
                    "start": 1493.08,
                    "end": 1493.8,
                    "word": " position",
                    "probability": 0.60302734375
                },
                {
                    "start": 1493.8,
                    "end": 1494.1,
                    "word": " of",
                    "probability": 0.908203125
                },
                {
                    "start": 1494.1,
                    "end": 1494.26,
                    "word": " an",
                    "probability": 0.904296875
                },
                {
                    "start": 1494.26,
                    "end": 1494.66,
                    "word": " attacker",
                    "probability": 0.78271484375
                },
                {
                    "start": 1494.66,
                    "end": 1495.12,
                    "word": ",",
                    "probability": 0.8525390625
                }
            ]
        },
        {
            "id": 225,
            "seek": 150869,
            "start": 1495.12,
            "end": 1501.56,
            "text": " you have to assume that browser agent is an unbounded space of potentially arbitrarily linked strings.",
            "tokens": [
                50914,
                291,
                362,
                281,
                6552,
                300,
                11185,
                9461,
                307,
                364,
                517,
                18767,
                292,
                1901,
                295,
                7263,
                19071,
                3289,
                9408,
                13985,
                13,
                51214
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19074876119594764,
            "compression_ratio": 1.6337448559670782,
            "no_speech_prob": 0.0025119781494140625,
            "words": [
                {
                    "start": 1495.12,
                    "end": 1495.28,
                    "word": " you",
                    "probability": 0.88623046875
                },
                {
                    "start": 1495.28,
                    "end": 1495.48,
                    "word": " have",
                    "probability": 0.8720703125
                },
                {
                    "start": 1495.48,
                    "end": 1495.6,
                    "word": " to",
                    "probability": 0.90673828125
                },
                {
                    "start": 1495.6,
                    "end": 1495.94,
                    "word": " assume",
                    "probability": 0.83447265625
                },
                {
                    "start": 1495.94,
                    "end": 1496.12,
                    "word": " that",
                    "probability": 0.84228515625
                },
                {
                    "start": 1496.12,
                    "end": 1496.38,
                    "word": " browser",
                    "probability": 0.476318359375
                },
                {
                    "start": 1496.38,
                    "end": 1496.72,
                    "word": " agent",
                    "probability": 0.65869140625
                },
                {
                    "start": 1496.72,
                    "end": 1497.0,
                    "word": " is",
                    "probability": 0.90234375
                },
                {
                    "start": 1497.0,
                    "end": 1497.2,
                    "word": " an",
                    "probability": 0.8974609375
                },
                {
                    "start": 1497.2,
                    "end": 1497.72,
                    "word": " unbounded",
                    "probability": 0.9010416666666666
                },
                {
                    "start": 1497.72,
                    "end": 1498.14,
                    "word": " space",
                    "probability": 0.81201171875
                },
                {
                    "start": 1498.14,
                    "end": 1498.5,
                    "word": " of",
                    "probability": 0.90380859375
                },
                {
                    "start": 1498.5,
                    "end": 1499.08,
                    "word": " potentially",
                    "probability": 0.64013671875
                },
                {
                    "start": 1499.08,
                    "end": 1500.0,
                    "word": " arbitrarily",
                    "probability": 0.81884765625
                },
                {
                    "start": 1500.0,
                    "end": 1500.36,
                    "word": " linked",
                    "probability": 0.67626953125
                },
                {
                    "start": 1500.36,
                    "end": 1500.98,
                    "word": " strings",
                    "probability": 0.81298828125
                },
                {
                    "start": 1500.98,
                    "end": 1501.56,
                    "word": ".",
                    "probability": 0.55810546875
                }
            ]
        },
        {
            "id": 226,
            "seek": 150869,
            "start": 1501.56,
            "end": 1506.98,
            "text": " Which means that if you see a value there in your logs,",
            "tokens": [
                51264,
                3013,
                1355,
                300,
                498,
                291,
                536,
                257,
                2158,
                456,
                294,
                428,
                20820,
                11,
                51464
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19074876119594764,
            "compression_ratio": 1.6337448559670782,
            "no_speech_prob": 0.0025119781494140625,
            "words": [
                {
                    "start": 1501.56,
                    "end": 1501.88,
                    "word": " Which",
                    "probability": 0.82373046875
                },
                {
                    "start": 1501.88,
                    "end": 1502.58,
                    "word": " means",
                    "probability": 0.8837890625
                },
                {
                    "start": 1502.58,
                    "end": 1503.38,
                    "word": " that",
                    "probability": 0.802734375
                },
                {
                    "start": 1503.38,
                    "end": 1503.58,
                    "word": " if",
                    "probability": 0.87939453125
                },
                {
                    "start": 1503.58,
                    "end": 1503.72,
                    "word": " you",
                    "probability": 0.89697265625
                },
                {
                    "start": 1503.72,
                    "end": 1504.04,
                    "word": " see",
                    "probability": 0.81787109375
                },
                {
                    "start": 1504.26,
                    "end": 1504.56,
                    "word": " a value",
                    "probability": 0.9130859375
                },
                {
                    "start": 1504.56,
                    "end": 1505.0,
                    "word": " there",
                    "probability": 0.861328125
                },
                {
                    "start": 1505.0,
                    "end": 1505.64,
                    "word": " in",
                    "probability": 0.767578125
                },
                {
                    "start": 1505.64,
                    "end": 1505.82,
                    "word": " your",
                    "probability": 0.80908203125
                },
                {
                    "start": 1505.82,
                    "end": 1506.32,
                    "word": " logs",
                    "probability": 0.71044921875
                },
                {
                    "start": 1506.32,
                    "end": 1506.98,
                    "word": ",",
                    "probability": 0.86669921875
                }
            ]
        },
        {
            "id": 227,
            "seek": 150869,
            "start": 1507.58,
            "end": 1511.2,
            "text": " then at least one user must have registered that.",
            "tokens": [
                51514,
                550,
                412,
                1935,
                472,
                4195,
                1633,
                362,
                13968,
                300,
                13,
                51664
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19074876119594764,
            "compression_ratio": 1.6337448559670782,
            "no_speech_prob": 0.0025119781494140625,
            "words": [
                {
                    "start": 1507.58,
                    "end": 1508.18,
                    "word": " then",
                    "probability": 0.71630859375
                },
                {
                    "start": 1508.18,
                    "end": 1508.46,
                    "word": " at",
                    "probability": 0.90771484375
                },
                {
                    "start": 1508.46,
                    "end": 1508.76,
                    "word": " least",
                    "probability": 0.8583984375
                },
                {
                    "start": 1508.76,
                    "end": 1509.04,
                    "word": " one",
                    "probability": 0.80859375
                },
                {
                    "start": 1509.04,
                    "end": 1509.34,
                    "word": " user",
                    "probability": 0.81103515625
                },
                {
                    "start": 1509.34,
                    "end": 1509.68,
                    "word": " must",
                    "probability": 0.806640625
                },
                {
                    "start": 1509.68,
                    "end": 1509.86,
                    "word": " have",
                    "probability": 0.8759765625
                },
                {
                    "start": 1509.86,
                    "end": 1510.24,
                    "word": " registered",
                    "probability": 0.68212890625
                },
                {
                    "start": 1510.24,
                    "end": 1510.64,
                    "word": " that",
                    "probability": 0.88525390625
                },
                {
                    "start": 1510.64,
                    "end": 1511.2,
                    "word": ".",
                    "probability": 0.98974609375
                }
            ]
        },
        {
            "id": 228,
            "seek": 150869,
            "start": 1511.2,
            "end": 1513.06,
            "text": " There are lots of ways of dealing with this.",
            "tokens": [
                51714,
                821,
                366,
                3195,
                295,
                2098,
                295,
                6260,
                365,
                341,
                13,
                51764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19074876119594764,
            "compression_ratio": 1.6337448559670782,
            "no_speech_prob": 0.0025119781494140625,
            "words": [
                {
                    "start": 1511.2,
                    "end": 1511.64,
                    "word": " There",
                    "probability": 0.79345703125
                },
                {
                    "start": 1511.64,
                    "end": 1511.72,
                    "word": " are",
                    "probability": 0.890625
                },
                {
                    "start": 1511.72,
                    "end": 1511.9,
                    "word": " lots",
                    "probability": 0.80078125
                },
                {
                    "start": 1511.9,
                    "end": 1512.02,
                    "word": " of",
                    "probability": 0.90673828125
                },
                {
                    "start": 1512.02,
                    "end": 1512.16,
                    "word": " ways",
                    "probability": 0.8046875
                },
                {
                    "start": 1512.16,
                    "end": 1512.3,
                    "word": " of",
                    "probability": 0.90380859375
                },
                {
                    "start": 1512.3,
                    "end": 1512.48,
                    "word": " dealing",
                    "probability": 0.8017578125
                },
                {
                    "start": 1512.48,
                    "end": 1512.72,
                    "word": " with",
                    "probability": 0.814453125
                },
                {
                    "start": 1512.72,
                    "end": 1512.98,
                    "word": " this",
                    "probability": 0.87939453125
                },
                {
                    "start": 1512.98,
                    "end": 1513.06,
                    "word": ".",
                    "probability": 0.82080078125
                }
            ]
        },
        {
            "id": 229,
            "seek": 153479,
            "start": 1513.06,
            "end": 1515.26,
            "text": " You could generate a bunch of random strings,",
            "tokens": [
                50364,
                509,
                727,
                8460,
                257,
                3840,
                295,
                4974,
                13985,
                11,
                50414
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1992590255344037,
            "compression_ratio": 1.6099585062240664,
            "no_speech_prob": 0.018829345703125,
            "words": [
                {
                    "start": 1513.06,
                    "end": 1513.12,
                    "word": " You",
                    "probability": 0.1163330078125
                },
                {
                    "start": 1513.12,
                    "end": 1513.32,
                    "word": " could",
                    "probability": 0.7578125
                },
                {
                    "start": 1513.32,
                    "end": 1513.68,
                    "word": " generate",
                    "probability": 0.79248046875
                },
                {
                    "start": 1513.8,
                    "end": 1513.92,
                    "word": " a bunch",
                    "probability": 0.81982421875
                },
                {
                    "start": 1513.92,
                    "end": 1514.02,
                    "word": " of",
                    "probability": 0.90283203125
                },
                {
                    "start": 1514.02,
                    "end": 1514.28,
                    "word": " random",
                    "probability": 0.76171875
                },
                {
                    "start": 1514.28,
                    "end": 1514.78,
                    "word": " strings",
                    "probability": 0.83349609375
                },
                {
                    "start": 1514.78,
                    "end": 1515.26,
                    "word": ",",
                    "probability": 0.1468505859375
                }
            ]
        },
        {
            "id": 230,
            "seek": 153479,
            "start": 1515.26,
            "end": 1520.24,
            "text": " and sort of give essentially zero noise to each of those.",
            "tokens": [
                50464,
                293,
                1333,
                295,
                976,
                4476,
                4018,
                5658,
                281,
                1184,
                295,
                729,
                13,
                50664
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1992590255344037,
            "compression_ratio": 1.6099585062240664,
            "no_speech_prob": 0.018829345703125,
            "words": [
                {
                    "start": 1515.26,
                    "end": 1516.06,
                    "word": " and",
                    "probability": 0.83544921875
                },
                {
                    "start": 1516.06,
                    "end": 1516.5,
                    "word": " sort",
                    "probability": 0.41845703125
                },
                {
                    "start": 1516.5,
                    "end": 1516.76,
                    "word": " of",
                    "probability": 0.904296875
                },
                {
                    "start": 1516.76,
                    "end": 1517.38,
                    "word": " give",
                    "probability": 0.7880859375
                },
                {
                    "start": 1517.38,
                    "end": 1518.28,
                    "word": " essentially",
                    "probability": 0.61962890625
                },
                {
                    "start": 1518.28,
                    "end": 1518.56,
                    "word": " zero",
                    "probability": 0.806640625
                },
                {
                    "start": 1518.56,
                    "end": 1518.88,
                    "word": " noise",
                    "probability": 0.8623046875
                },
                {
                    "start": 1518.88,
                    "end": 1519.06,
                    "word": " to",
                    "probability": 0.91015625
                },
                {
                    "start": 1519.06,
                    "end": 1519.3,
                    "word": " each",
                    "probability": 0.8603515625
                },
                {
                    "start": 1519.3,
                    "end": 1519.44,
                    "word": " of",
                    "probability": 0.900390625
                },
                {
                    "start": 1519.44,
                    "end": 1519.68,
                    "word": " those",
                    "probability": 0.8447265625
                },
                {
                    "start": 1519.68,
                    "end": 1520.24,
                    "word": ".",
                    "probability": 0.63232421875
                }
            ]
        },
        {
            "id": 231,
            "seek": 153479,
            "start": 1520.42,
            "end": 1525.6,
            "text": " But you, for an unbounded space, would have to generate unboundedly many browser agents.",
            "tokens": [
                50714,
                583,
                291,
                11,
                337,
                364,
                517,
                18767,
                292,
                1901,
                11,
                576,
                362,
                281,
                8460,
                517,
                18767,
                13516,
                867,
                11185,
                12554,
                13,
                50914
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1992590255344037,
            "compression_ratio": 1.6099585062240664,
            "no_speech_prob": 0.018829345703125,
            "words": [
                {
                    "start": 1520.42,
                    "end": 1520.86,
                    "word": " But",
                    "probability": 0.83984375
                },
                {
                    "start": 1520.86,
                    "end": 1521.04,
                    "word": " you",
                    "probability": 0.5546875
                },
                {
                    "start": 1521.04,
                    "end": 1521.38,
                    "word": ",",
                    "probability": 0.76953125
                },
                {
                    "start": 1521.38,
                    "end": 1521.46,
                    "word": " for",
                    "probability": 0.8984375
                },
                {
                    "start": 1521.46,
                    "end": 1521.56,
                    "word": " an",
                    "probability": 0.89794921875
                },
                {
                    "start": 1521.56,
                    "end": 1521.94,
                    "word": " unbounded",
                    "probability": 0.9208984375
                },
                {
                    "start": 1521.94,
                    "end": 1522.22,
                    "word": " space",
                    "probability": 0.8310546875
                },
                {
                    "start": 1522.22,
                    "end": 1522.26,
                    "word": ",",
                    "probability": 0.87060546875
                },
                {
                    "start": 1522.26,
                    "end": 1522.38,
                    "word": " would",
                    "probability": 0.83984375
                },
                {
                    "start": 1522.38,
                    "end": 1522.52,
                    "word": " have",
                    "probability": 0.876953125
                },
                {
                    "start": 1522.52,
                    "end": 1522.88,
                    "word": " to",
                    "probability": 0.8994140625
                },
                {
                    "start": 1522.88,
                    "end": 1523.38,
                    "word": " generate",
                    "probability": 0.8359375
                },
                {
                    "start": 1523.38,
                    "end": 1524.04,
                    "word": " unboundedly",
                    "probability": 0.8948567708333334
                },
                {
                    "start": 1524.04,
                    "end": 1524.32,
                    "word": " many",
                    "probability": 0.8115234375
                },
                {
                    "start": 1524.32,
                    "end": 1524.66,
                    "word": " browser",
                    "probability": 0.7646484375
                },
                {
                    "start": 1524.66,
                    "end": 1525.06,
                    "word": " agents",
                    "probability": 0.7802734375
                },
                {
                    "start": 1525.06,
                    "end": 1525.6,
                    "word": ".",
                    "probability": 0.986328125
                }
            ]
        },
        {
            "id": 232,
            "seek": 153479,
            "start": 1525.6,
            "end": 1528.92,
            "text": " So the way that we do it is we basically just filter",
            "tokens": [
                50964,
                407,
                264,
                636,
                300,
                321,
                360,
                309,
                307,
                321,
                1936,
                445,
                6608,
                51114
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1992590255344037,
            "compression_ratio": 1.6099585062240664,
            "no_speech_prob": 0.018829345703125,
            "words": [
                {
                    "start": 1525.6,
                    "end": 1525.7,
                    "word": " So",
                    "probability": 0.7685546875
                },
                {
                    "start": 1525.7,
                    "end": 1525.8,
                    "word": " the",
                    "probability": 0.634765625
                },
                {
                    "start": 1525.8,
                    "end": 1525.9,
                    "word": " way",
                    "probability": 0.90380859375
                },
                {
                    "start": 1525.9,
                    "end": 1526.02,
                    "word": " that",
                    "probability": 0.837890625
                },
                {
                    "start": 1526.02,
                    "end": 1526.14,
                    "word": " we",
                    "probability": 0.90673828125
                },
                {
                    "start": 1526.14,
                    "end": 1526.34,
                    "word": " do",
                    "probability": 0.90673828125
                },
                {
                    "start": 1526.34,
                    "end": 1526.58,
                    "word": " it",
                    "probability": 0.90380859375
                },
                {
                    "start": 1526.58,
                    "end": 1527.34,
                    "word": " is",
                    "probability": 0.73583984375
                },
                {
                    "start": 1527.34,
                    "end": 1528.0,
                    "word": " we",
                    "probability": 0.59912109375
                },
                {
                    "start": 1528.0,
                    "end": 1528.32,
                    "word": " basically",
                    "probability": 0.90087890625
                },
                {
                    "start": 1528.32,
                    "end": 1528.6,
                    "word": " just",
                    "probability": 0.88720703125
                },
                {
                    "start": 1528.6,
                    "end": 1528.92,
                    "word": " filter",
                    "probability": 0.89306640625
                }
            ]
        },
        {
            "id": 233,
            "seek": 153479,
            "start": 1528.92,
            "end": 1535.32,
            "text": " and say that at least tau users need to contribute to any particular partition.",
            "tokens": [
                51164,
                293,
                584,
                300,
                412,
                1935,
                17842,
                5022,
                643,
                281,
                10586,
                281,
                604,
                1729,
                24808,
                13,
                51414
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1992590255344037,
            "compression_ratio": 1.6099585062240664,
            "no_speech_prob": 0.018829345703125,
            "words": [
                {
                    "start": 1528.92,
                    "end": 1529.18,
                    "word": " and",
                    "probability": 0.6572265625
                },
                {
                    "start": 1529.18,
                    "end": 1529.48,
                    "word": " say",
                    "probability": 0.89990234375
                },
                {
                    "start": 1529.48,
                    "end": 1530.08,
                    "word": " that",
                    "probability": 0.86572265625
                },
                {
                    "start": 1530.08,
                    "end": 1531.0,
                    "word": " at",
                    "probability": 0.83837890625
                },
                {
                    "start": 1531.0,
                    "end": 1531.6,
                    "word": " least",
                    "probability": 0.85595703125
                },
                {
                    "start": 1531.6,
                    "end": 1532.06,
                    "word": " tau",
                    "probability": 0.4560546875
                },
                {
                    "start": 1532.06,
                    "end": 1532.52,
                    "word": " users",
                    "probability": 0.82958984375
                },
                {
                    "start": 1532.52,
                    "end": 1532.82,
                    "word": " need",
                    "probability": 0.810546875
                },
                {
                    "start": 1532.82,
                    "end": 1532.96,
                    "word": " to",
                    "probability": 0.90283203125
                },
                {
                    "start": 1532.96,
                    "end": 1533.26,
                    "word": " contribute",
                    "probability": 0.87451171875
                },
                {
                    "start": 1533.26,
                    "end": 1533.48,
                    "word": " to",
                    "probability": 0.90576171875
                },
                {
                    "start": 1533.48,
                    "end": 1533.66,
                    "word": " any",
                    "probability": 0.81494140625
                },
                {
                    "start": 1533.66,
                    "end": 1534.12,
                    "word": " particular",
                    "probability": 0.8251953125
                },
                {
                    "start": 1534.12,
                    "end": 1534.76,
                    "word": " partition",
                    "probability": 0.9423828125
                },
                {
                    "start": 1534.76,
                    "end": 1535.32,
                    "word": ".",
                    "probability": 0.98974609375
                }
            ]
        },
        {
            "id": 234,
            "seek": 153479,
            "start": 1535.5,
            "end": 1539.16,
            "text": " It turns out that there's a direct line between tau and delta.",
            "tokens": [
                51464,
                467,
                4523,
                484,
                300,
                456,
                311,
                257,
                2047,
                1622,
                1296,
                17842,
                293,
                8289,
                13,
                51614
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1992590255344037,
            "compression_ratio": 1.6099585062240664,
            "no_speech_prob": 0.018829345703125,
            "words": [
                {
                    "start": 1535.5,
                    "end": 1535.74,
                    "word": " It",
                    "probability": 0.89306640625
                },
                {
                    "start": 1535.74,
                    "end": 1535.92,
                    "word": " turns",
                    "probability": 0.7646484375
                },
                {
                    "start": 1535.92,
                    "end": 1536.18,
                    "word": " out",
                    "probability": 0.8203125
                },
                {
                    "start": 1536.18,
                    "end": 1536.56,
                    "word": " that",
                    "probability": 0.8681640625
                },
                {
                    "start": 1536.56,
                    "end": 1536.8,
                    "word": " there's",
                    "probability": 0.894775390625
                },
                {
                    "start": 1536.92,
                    "end": 1537.18,
                    "word": " a direct",
                    "probability": 0.8125
                },
                {
                    "start": 1537.18,
                    "end": 1537.52,
                    "word": " line",
                    "probability": 0.8564453125
                },
                {
                    "start": 1537.52,
                    "end": 1537.86,
                    "word": " between",
                    "probability": 0.78076171875
                },
                {
                    "start": 1537.86,
                    "end": 1538.2,
                    "word": " tau",
                    "probability": 0.80517578125
                },
                {
                    "start": 1538.2,
                    "end": 1538.34,
                    "word": " and",
                    "probability": 0.89501953125
                },
                {
                    "start": 1538.34,
                    "end": 1538.6,
                    "word": " delta",
                    "probability": 0.84912109375
                },
                {
                    "start": 1538.6,
                    "end": 1539.16,
                    "word": ".",
                    "probability": 0.465087890625
                }
            ]
        },
        {
            "id": 235,
            "seek": 156341,
            "start": 1539.16,
            "end": 1543.6,
            "text": " So it's really easy to reason about what sort of the correct tau or the correct delta",
            "tokens": [
                50364,
                407,
                309,
                311,
                534,
                1858,
                281,
                1778,
                466,
                437,
                1333,
                295,
                264,
                3006,
                17842,
                420,
                264,
                3006,
                8289,
                50564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.16486565977613502,
            "compression_ratio": 1.6015936254980079,
            "no_speech_prob": 0.006191253662109375,
            "words": [
                {
                    "start": 1539.16,
                    "end": 1539.46,
                    "word": " So",
                    "probability": 0.445068359375
                },
                {
                    "start": 1539.46,
                    "end": 1539.6,
                    "word": " it's",
                    "probability": 0.805908203125
                },
                {
                    "start": 1539.6,
                    "end": 1539.84,
                    "word": " really",
                    "probability": 0.77734375
                },
                {
                    "start": 1539.84,
                    "end": 1540.16,
                    "word": " easy",
                    "probability": 0.81201171875
                },
                {
                    "start": 1540.16,
                    "end": 1540.36,
                    "word": " to",
                    "probability": 0.89794921875
                },
                {
                    "start": 1540.36,
                    "end": 1540.58,
                    "word": " reason",
                    "probability": 0.90625
                },
                {
                    "start": 1540.58,
                    "end": 1540.84,
                    "word": " about",
                    "probability": 0.82421875
                },
                {
                    "start": 1540.84,
                    "end": 1541.14,
                    "word": " what",
                    "probability": 0.88818359375
                },
                {
                    "start": 1541.14,
                    "end": 1541.56,
                    "word": " sort",
                    "probability": 0.459716796875
                },
                {
                    "start": 1541.56,
                    "end": 1541.66,
                    "word": " of",
                    "probability": 0.90283203125
                },
                {
                    "start": 1541.66,
                    "end": 1541.88,
                    "word": " the",
                    "probability": 0.79638671875
                },
                {
                    "start": 1541.88,
                    "end": 1542.2,
                    "word": " correct",
                    "probability": 0.8876953125
                },
                {
                    "start": 1542.2,
                    "end": 1542.64,
                    "word": " tau",
                    "probability": 0.703125
                },
                {
                    "start": 1542.64,
                    "end": 1542.86,
                    "word": " or",
                    "probability": 0.83740234375
                },
                {
                    "start": 1542.86,
                    "end": 1543.0,
                    "word": " the",
                    "probability": 0.79248046875
                },
                {
                    "start": 1543.0,
                    "end": 1543.18,
                    "word": " correct",
                    "probability": 0.8740234375
                },
                {
                    "start": 1543.18,
                    "end": 1543.6,
                    "word": " delta",
                    "probability": 0.8095703125
                }
            ]
        },
        {
            "id": 236,
            "seek": 156341,
            "start": 1544.16,
            "end": 1546.86,
            "text": " needs to be in terms of the existing parameters.",
            "tokens": [
                50614,
                2203,
                281,
                312,
                294,
                2115,
                295,
                264,
                6741,
                9834,
                13,
                50714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.16486565977613502,
            "compression_ratio": 1.6015936254980079,
            "no_speech_prob": 0.006191253662109375,
            "words": [
                {
                    "start": 1544.16,
                    "end": 1544.6,
                    "word": " needs",
                    "probability": 0.71240234375
                },
                {
                    "start": 1544.6,
                    "end": 1544.74,
                    "word": " to",
                    "probability": 0.904296875
                },
                {
                    "start": 1544.74,
                    "end": 1544.88,
                    "word": " be",
                    "probability": 0.904296875
                },
                {
                    "start": 1544.88,
                    "end": 1545.0,
                    "word": " in",
                    "probability": 0.88916015625
                },
                {
                    "start": 1545.0,
                    "end": 1545.22,
                    "word": " terms",
                    "probability": 0.78564453125
                },
                {
                    "start": 1545.22,
                    "end": 1545.34,
                    "word": " of",
                    "probability": 0.90673828125
                },
                {
                    "start": 1545.34,
                    "end": 1545.46,
                    "word": " the",
                    "probability": 0.8095703125
                },
                {
                    "start": 1545.46,
                    "end": 1545.76,
                    "word": " existing",
                    "probability": 0.83935546875
                },
                {
                    "start": 1545.76,
                    "end": 1546.3,
                    "word": " parameters",
                    "probability": 0.892578125
                },
                {
                    "start": 1546.3,
                    "end": 1546.86,
                    "word": ".",
                    "probability": 0.97802734375
                }
            ]
        },
        {
            "id": 237,
            "seek": 156341,
            "start": 1547.94,
            "end": 1552.34,
            "text": " As I said earlier, the nature of the noise is a decaying exponential.",
            "tokens": [
                50764,
                1018,
                286,
                848,
                3071,
                11,
                264,
                3687,
                295,
                264,
                5658,
                307,
                257,
                21039,
                278,
                21510,
                13,
                50964
            ],
            "temperature": 0.0,
            "avg_logprob": -0.16486565977613502,
            "compression_ratio": 1.6015936254980079,
            "no_speech_prob": 0.006191253662109375,
            "words": [
                {
                    "start": 1547.94,
                    "end": 1548.5,
                    "word": " As",
                    "probability": 0.89404296875
                },
                {
                    "start": 1548.5,
                    "end": 1548.62,
                    "word": " I",
                    "probability": 0.994140625
                },
                {
                    "start": 1548.62,
                    "end": 1548.76,
                    "word": " said",
                    "probability": 0.8779296875
                },
                {
                    "start": 1548.76,
                    "end": 1549.1,
                    "word": " earlier",
                    "probability": 0.90673828125
                },
                {
                    "start": 1549.1,
                    "end": 1549.64,
                    "word": ",",
                    "probability": 0.92578125
                },
                {
                    "start": 1549.64,
                    "end": 1549.86,
                    "word": " the",
                    "probability": 0.80615234375
                },
                {
                    "start": 1549.86,
                    "end": 1550.16,
                    "word": " nature",
                    "probability": 0.78759765625
                },
                {
                    "start": 1550.16,
                    "end": 1550.32,
                    "word": " of",
                    "probability": 0.904296875
                },
                {
                    "start": 1550.32,
                    "end": 1550.44,
                    "word": " the",
                    "probability": 0.81103515625
                },
                {
                    "start": 1550.44,
                    "end": 1550.68,
                    "word": " noise",
                    "probability": 0.89013671875
                },
                {
                    "start": 1550.68,
                    "end": 1550.88,
                    "word": " is",
                    "probability": 0.90771484375
                },
                {
                    "start": 1551.02,
                    "end": 1551.4,
                    "word": " a decaying",
                    "probability": 0.845458984375
                },
                {
                    "start": 1551.4,
                    "end": 1551.82,
                    "word": " exponential",
                    "probability": 0.904296875
                },
                {
                    "start": 1551.82,
                    "end": 1552.34,
                    "word": ".",
                    "probability": 0.7265625
                }
            ]
        },
        {
            "id": 238,
            "seek": 156341,
            "start": 1552.34,
            "end": 1555.18,
            "text": " So even for 50 users or 100 users,",
            "tokens": [
                51014,
                407,
                754,
                337,
                2625,
                5022,
                420,
                2319,
                5022,
                11,
                51114
            ],
            "temperature": 0.0,
            "avg_logprob": -0.16486565977613502,
            "compression_ratio": 1.6015936254980079,
            "no_speech_prob": 0.006191253662109375,
            "words": [
                {
                    "start": 1552.34,
                    "end": 1552.52,
                    "word": " So",
                    "probability": 0.896484375
                },
                {
                    "start": 1552.52,
                    "end": 1552.88,
                    "word": " even",
                    "probability": 0.7685546875
                },
                {
                    "start": 1552.88,
                    "end": 1553.1,
                    "word": " for",
                    "probability": 0.89794921875
                },
                {
                    "start": 1553.1,
                    "end": 1553.42,
                    "word": " 50",
                    "probability": 0.8828125
                },
                {
                    "start": 1553.42,
                    "end": 1553.8,
                    "word": " users",
                    "probability": 0.83447265625
                },
                {
                    "start": 1553.8,
                    "end": 1553.98,
                    "word": " or",
                    "probability": 0.876953125
                },
                {
                    "start": 1553.98,
                    "end": 1554.24,
                    "word": " 100",
                    "probability": 0.6328125
                },
                {
                    "start": 1554.24,
                    "end": 1554.72,
                    "word": " users",
                    "probability": 0.8515625
                },
                {
                    "start": 1554.72,
                    "end": 1555.18,
                    "word": ",",
                    "probability": 0.92333984375
                }
            ]
        },
        {
            "id": 239,
            "seek": 156341,
            "start": 1555.18,
            "end": 1559.34,
            "text": " you might be at a vanishingly small, perhaps unmeasurably small,",
            "tokens": [
                51164,
                291,
                1062,
                312,
                412,
                257,
                3161,
                3807,
                356,
                1359,
                11,
                4317,
                517,
                1398,
                296,
                374,
                1188,
                1359,
                11,
                51314
            ],
            "temperature": 0.0,
            "avg_logprob": -0.16486565977613502,
            "compression_ratio": 1.6015936254980079,
            "no_speech_prob": 0.006191253662109375,
            "words": [
                {
                    "start": 1555.18,
                    "end": 1555.46,
                    "word": " you",
                    "probability": 0.8935546875
                },
                {
                    "start": 1555.46,
                    "end": 1555.9,
                    "word": " might",
                    "probability": 0.7685546875
                },
                {
                    "start": 1555.9,
                    "end": 1556.04,
                    "word": " be",
                    "probability": 0.904296875
                },
                {
                    "start": 1556.04,
                    "end": 1556.16,
                    "word": " at",
                    "probability": 0.904296875
                },
                {
                    "start": 1556.34,
                    "end": 1557.08,
                    "word": " a vanishingly",
                    "probability": 0.9371744791666666
                },
                {
                    "start": 1557.08,
                    "end": 1557.42,
                    "word": " small",
                    "probability": 0.84716796875
                },
                {
                    "start": 1557.42,
                    "end": 1557.56,
                    "word": ",",
                    "probability": 0.93310546875
                },
                {
                    "start": 1557.56,
                    "end": 1557.76,
                    "word": " perhaps",
                    "probability": 0.84521484375
                },
                {
                    "start": 1557.76,
                    "end": 1558.4,
                    "word": " unmeasurably",
                    "probability": 0.9380859375
                },
                {
                    "start": 1558.4,
                    "end": 1558.94,
                    "word": " small",
                    "probability": 0.8466796875
                },
                {
                    "start": 1558.94,
                    "end": 1559.34,
                    "word": ",",
                    "probability": 0.34326171875
                }
            ]
        },
        {
            "id": 240,
            "seek": 156341,
            "start": 1559.34,
            "end": 1564.76,
            "text": " probability that a particular partition is surfaced incorrectly.",
            "tokens": [
                51364,
                8482,
                300,
                257,
                1729,
                24808,
                307,
                9684,
                3839,
                42892,
                13,
                51614
            ],
            "temperature": 0.0,
            "avg_logprob": -0.16486565977613502,
            "compression_ratio": 1.6015936254980079,
            "no_speech_prob": 0.006191253662109375,
            "words": [
                {
                    "start": 1559.34,
                    "end": 1560.26,
                    "word": " probability",
                    "probability": 0.9140625
                },
                {
                    "start": 1560.26,
                    "end": 1561.36,
                    "word": " that",
                    "probability": 0.87255859375
                },
                {
                    "start": 1562.3,
                    "end": 1562.58,
                    "word": " a particular",
                    "probability": 0.81396484375
                },
                {
                    "start": 1562.58,
                    "end": 1563.02,
                    "word": " partition",
                    "probability": 0.9375
                },
                {
                    "start": 1563.02,
                    "end": 1563.3,
                    "word": " is",
                    "probability": 0.6953125
                },
                {
                    "start": 1563.3,
                    "end": 1563.66,
                    "word": " surfaced",
                    "probability": 0.826904296875
                },
                {
                    "start": 1563.66,
                    "end": 1564.2,
                    "word": " incorrectly",
                    "probability": 0.87890625
                },
                {
                    "start": 1564.2,
                    "end": 1564.76,
                    "word": ".",
                    "probability": 0.99462890625
                }
            ]
        },
        {
            "id": 241,
            "seek": 156341,
            "start": 1565.78,
            "end": 1567.78,
            "text": " So you think, we're done, right?",
            "tokens": [
                51664,
                407,
                291,
                519,
                11,
                321,
                434,
                1096,
                11,
                558,
                30,
                51764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.16486565977613502,
            "compression_ratio": 1.6015936254980079,
            "no_speech_prob": 0.006191253662109375,
            "words": [
                {
                    "start": 1565.78,
                    "end": 1566.34,
                    "word": " So",
                    "probability": 0.89697265625
                },
                {
                    "start": 1566.34,
                    "end": 1566.44,
                    "word": " you",
                    "probability": 0.87890625
                },
                {
                    "start": 1566.44,
                    "end": 1566.7,
                    "word": " think",
                    "probability": 0.681640625
                },
                {
                    "start": 1566.7,
                    "end": 1566.8,
                    "word": ",",
                    "probability": 0.837890625
                },
                {
                    "start": 1566.8,
                    "end": 1567.2,
                    "word": " we're",
                    "probability": 0.54095458984375
                },
                {
                    "start": 1567.2,
                    "end": 1567.36,
                    "word": " done",
                    "probability": 0.7958984375
                },
                {
                    "start": 1567.36,
                    "end": 1567.42,
                    "word": ",",
                    "probability": 0.85302734375
                },
                {
                    "start": 1567.42,
                    "end": 1567.58,
                    "word": " right",
                    "probability": 0.88427734375
                },
                {
                    "start": 1567.58,
                    "end": 1567.78,
                    "word": "?",
                    "probability": 0.96435546875
                }
            ]
        },
        {
            "id": 242,
            "seek": 158939,
            "start": 1567.78,
            "end": 1571.36,
            "text": " Well, almost. There's a little bit more nuance to this.",
            "tokens": [
                50364,
                1042,
                11,
                1920,
                13,
                821,
                311,
                257,
                707,
                857,
                544,
                42625,
                281,
                341,
                13,
                50514
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17571272060536502,
            "compression_ratio": 1.6125461254612545,
            "no_speech_prob": 0.0014095306396484375,
            "words": [
                {
                    "start": 1567.78,
                    "end": 1568.06,
                    "word": " Well",
                    "probability": 0.46337890625
                },
                {
                    "start": 1568.06,
                    "end": 1568.22,
                    "word": ",",
                    "probability": 0.76708984375
                },
                {
                    "start": 1568.22,
                    "end": 1568.7,
                    "word": " almost",
                    "probability": 0.67822265625
                },
                {
                    "start": 1568.7,
                    "end": 1568.84,
                    "word": ".",
                    "probability": 0.69921875
                },
                {
                    "start": 1568.84,
                    "end": 1569.06,
                    "word": " There's",
                    "probability": 0.85693359375
                },
                {
                    "start": 1569.24,
                    "end": 1569.42,
                    "word": " a little",
                    "probability": 0.80810546875
                },
                {
                    "start": 1569.42,
                    "end": 1569.74,
                    "word": " bit",
                    "probability": 0.89111328125
                },
                {
                    "start": 1569.74,
                    "end": 1570.0,
                    "word": " more",
                    "probability": 0.88623046875
                },
                {
                    "start": 1570.0,
                    "end": 1570.38,
                    "word": " nuance",
                    "probability": 0.9130859375
                },
                {
                    "start": 1570.38,
                    "end": 1570.56,
                    "word": " to",
                    "probability": 0.9033203125
                },
                {
                    "start": 1570.56,
                    "end": 1570.92,
                    "word": " this",
                    "probability": 0.8837890625
                },
                {
                    "start": 1570.92,
                    "end": 1571.36,
                    "word": ".",
                    "probability": 0.48779296875
                }
            ]
        },
        {
            "id": 243,
            "seek": 158939,
            "start": 1571.36,
            "end": 1576.58,
            "text": " And this is kind of one of the big strengths of this type of a privacy model in SQL",
            "tokens": [
                50564,
                400,
                341,
                307,
                733,
                295,
                472,
                295,
                264,
                955,
                16986,
                295,
                341,
                2010,
                295,
                257,
                11427,
                2316,
                294,
                19200,
                50764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17571272060536502,
            "compression_ratio": 1.6125461254612545,
            "no_speech_prob": 0.0014095306396484375,
            "words": [
                {
                    "start": 1571.36,
                    "end": 1571.5,
                    "word": " And",
                    "probability": 0.8193359375
                },
                {
                    "start": 1571.5,
                    "end": 1571.66,
                    "word": " this",
                    "probability": 0.87109375
                },
                {
                    "start": 1571.66,
                    "end": 1571.84,
                    "word": " is",
                    "probability": 0.91064453125
                },
                {
                    "start": 1571.84,
                    "end": 1572.0,
                    "word": " kind",
                    "probability": 0.65478515625
                },
                {
                    "start": 1572.0,
                    "end": 1572.12,
                    "word": " of",
                    "probability": 0.9091796875
                },
                {
                    "start": 1572.12,
                    "end": 1572.38,
                    "word": " one",
                    "probability": 0.81201171875
                },
                {
                    "start": 1572.38,
                    "end": 1572.6,
                    "word": " of",
                    "probability": 0.900390625
                },
                {
                    "start": 1572.6,
                    "end": 1573.26,
                    "word": " the",
                    "probability": 0.81298828125
                },
                {
                    "start": 1573.26,
                    "end": 1574.1,
                    "word": " big",
                    "probability": 0.78662109375
                },
                {
                    "start": 1574.1,
                    "end": 1574.4,
                    "word": " strengths",
                    "probability": 0.6484375
                },
                {
                    "start": 1574.4,
                    "end": 1574.64,
                    "word": " of",
                    "probability": 0.90380859375
                },
                {
                    "start": 1574.64,
                    "end": 1575.0,
                    "word": " this",
                    "probability": 0.85107421875
                },
                {
                    "start": 1575.0,
                    "end": 1575.3,
                    "word": " type",
                    "probability": 0.9306640625
                },
                {
                    "start": 1575.3,
                    "end": 1575.42,
                    "word": " of",
                    "probability": 0.90087890625
                },
                {
                    "start": 1575.56,
                    "end": 1575.8,
                    "word": " a privacy",
                    "probability": 0.806640625
                },
                {
                    "start": 1575.8,
                    "end": 1576.16,
                    "word": " model",
                    "probability": 0.82763671875
                },
                {
                    "start": 1576.16,
                    "end": 1576.34,
                    "word": " in",
                    "probability": 0.7490234375
                },
                {
                    "start": 1576.34,
                    "end": 1576.58,
                    "word": " SQL",
                    "probability": 0.9287109375
                }
            ]
        },
        {
            "id": 244,
            "seek": 158939,
            "start": 1576.58,
            "end": 1578.18,
            "text": " is that you can enforce it.",
            "tokens": [
                50814,
                307,
                300,
                291,
                393,
                24825,
                309,
                13,
                50864
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17571272060536502,
            "compression_ratio": 1.6125461254612545,
            "no_speech_prob": 0.0014095306396484375,
            "words": [
                {
                    "start": 1576.58,
                    "end": 1576.8,
                    "word": " is",
                    "probability": 0.70361328125
                },
                {
                    "start": 1576.8,
                    "end": 1576.9,
                    "word": " that",
                    "probability": 0.88232421875
                },
                {
                    "start": 1576.9,
                    "end": 1577.0,
                    "word": " you",
                    "probability": 0.8935546875
                },
                {
                    "start": 1577.0,
                    "end": 1577.18,
                    "word": " can",
                    "probability": 0.89453125
                },
                {
                    "start": 1577.18,
                    "end": 1577.46,
                    "word": " enforce",
                    "probability": 0.744140625
                },
                {
                    "start": 1577.46,
                    "end": 1577.7,
                    "word": " it",
                    "probability": 0.90673828125
                },
                {
                    "start": 1577.7,
                    "end": 1578.18,
                    "word": ".",
                    "probability": 0.99072265625
                }
            ]
        },
        {
            "id": 245,
            "seek": 158939,
            "start": 1578.9,
            "end": 1583.8,
            "text": " So an individual user in any particular session might be using a single browser.",
            "tokens": [
                50914,
                407,
                364,
                2609,
                4195,
                294,
                604,
                1729,
                5481,
                1062,
                312,
                1228,
                257,
                2167,
                11185,
                13,
                51114
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17571272060536502,
            "compression_ratio": 1.6125461254612545,
            "no_speech_prob": 0.0014095306396484375,
            "words": [
                {
                    "start": 1578.9,
                    "end": 1579.38,
                    "word": " So",
                    "probability": 0.87548828125
                },
                {
                    "start": 1579.38,
                    "end": 1579.86,
                    "word": " an",
                    "probability": 0.5712890625
                },
                {
                    "start": 1579.86,
                    "end": 1580.2,
                    "word": " individual",
                    "probability": 0.72998046875
                },
                {
                    "start": 1580.2,
                    "end": 1580.86,
                    "word": " user",
                    "probability": 0.81689453125
                },
                {
                    "start": 1580.86,
                    "end": 1581.4,
                    "word": " in",
                    "probability": 0.81298828125
                },
                {
                    "start": 1581.4,
                    "end": 1581.6,
                    "word": " any",
                    "probability": 0.82568359375
                },
                {
                    "start": 1581.6,
                    "end": 1581.94,
                    "word": " particular",
                    "probability": 0.82666015625
                },
                {
                    "start": 1581.94,
                    "end": 1582.32,
                    "word": " session",
                    "probability": 0.841796875
                },
                {
                    "start": 1582.32,
                    "end": 1582.56,
                    "word": " might",
                    "probability": 0.8017578125
                },
                {
                    "start": 1582.56,
                    "end": 1582.68,
                    "word": " be",
                    "probability": 0.904296875
                },
                {
                    "start": 1582.68,
                    "end": 1582.88,
                    "word": " using",
                    "probability": 0.82958984375
                },
                {
                    "start": 1583.04,
                    "end": 1583.24,
                    "word": " a single",
                    "probability": 0.88330078125
                },
                {
                    "start": 1583.24,
                    "end": 1583.64,
                    "word": " browser",
                    "probability": 0.72509765625
                },
                {
                    "start": 1583.64,
                    "end": 1583.8,
                    "word": ".",
                    "probability": 0.349609375
                }
            ]
        },
        {
            "id": 246,
            "seek": 158939,
            "start": 1583.8,
            "end": 1586.6,
            "text": " But across all their sessions, they may have used many browsers,",
            "tokens": [
                51164,
                583,
                2108,
                439,
                641,
                11081,
                11,
                436,
                815,
                362,
                1143,
                867,
                36069,
                11,
                51264
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17571272060536502,
            "compression_ratio": 1.6125461254612545,
            "no_speech_prob": 0.0014095306396484375,
            "words": [
                {
                    "start": 1583.8,
                    "end": 1584.08,
                    "word": " But",
                    "probability": 0.890625
                },
                {
                    "start": 1584.08,
                    "end": 1584.3,
                    "word": " across",
                    "probability": 0.71630859375
                },
                {
                    "start": 1584.3,
                    "end": 1584.56,
                    "word": " all",
                    "probability": 0.8994140625
                },
                {
                    "start": 1584.56,
                    "end": 1584.7,
                    "word": " their",
                    "probability": 0.87109375
                },
                {
                    "start": 1584.7,
                    "end": 1585.06,
                    "word": " sessions",
                    "probability": 0.80908203125
                },
                {
                    "start": 1585.06,
                    "end": 1585.18,
                    "word": ",",
                    "probability": 0.84521484375
                },
                {
                    "start": 1585.18,
                    "end": 1585.28,
                    "word": " they",
                    "probability": 0.7353515625
                },
                {
                    "start": 1585.28,
                    "end": 1585.42,
                    "word": " may",
                    "probability": 0.9033203125
                },
                {
                    "start": 1585.42,
                    "end": 1585.54,
                    "word": " have",
                    "probability": 0.88037109375
                },
                {
                    "start": 1585.54,
                    "end": 1585.74,
                    "word": " used",
                    "probability": 0.79443359375
                },
                {
                    "start": 1585.74,
                    "end": 1585.98,
                    "word": " many",
                    "probability": 0.8291015625
                },
                {
                    "start": 1585.98,
                    "end": 1586.42,
                    "word": " browsers",
                    "probability": 0.80615234375
                },
                {
                    "start": 1586.42,
                    "end": 1586.6,
                    "word": ",",
                    "probability": 0.77978515625
                }
            ]
        },
        {
            "id": 247,
            "seek": 158939,
            "start": 1586.6,
            "end": 1588.3,
            "text": " unboundedly many, in fact.",
            "tokens": [
                51314,
                517,
                18767,
                13516,
                867,
                11,
                294,
                1186,
                13,
                51364
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17571272060536502,
            "compression_ratio": 1.6125461254612545,
            "no_speech_prob": 0.0014095306396484375,
            "words": [
                {
                    "start": 1586.6,
                    "end": 1587.1,
                    "word": " unboundedly",
                    "probability": 0.8600260416666666
                },
                {
                    "start": 1587.1,
                    "end": 1587.38,
                    "word": " many",
                    "probability": 0.78759765625
                },
                {
                    "start": 1587.38,
                    "end": 1587.46,
                    "word": ",",
                    "probability": 0.6337890625
                },
                {
                    "start": 1587.46,
                    "end": 1587.54,
                    "word": " in",
                    "probability": 0.900390625
                },
                {
                    "start": 1587.54,
                    "end": 1587.82,
                    "word": " fact",
                    "probability": 0.8076171875
                },
                {
                    "start": 1587.82,
                    "end": 1588.3,
                    "word": ".",
                    "probability": 0.9521484375
                }
            ]
        },
        {
            "id": 248,
            "seek": 158939,
            "start": 1588.36,
            "end": 1590.48,
            "text": " And from the perspective of differential privacy,",
            "tokens": [
                51414,
                400,
                490,
                264,
                4585,
                295,
                15756,
                11427,
                11,
                51464
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17571272060536502,
            "compression_ratio": 1.6125461254612545,
            "no_speech_prob": 0.0014095306396484375,
            "words": [
                {
                    "start": 1588.36,
                    "end": 1588.72,
                    "word": " And",
                    "probability": 0.88232421875
                },
                {
                    "start": 1588.72,
                    "end": 1588.92,
                    "word": " from",
                    "probability": 0.810546875
                },
                {
                    "start": 1588.92,
                    "end": 1589.08,
                    "word": " the",
                    "probability": 0.82080078125
                },
                {
                    "start": 1589.08,
                    "end": 1589.48,
                    "word": " perspective",
                    "probability": 0.791015625
                },
                {
                    "start": 1589.48,
                    "end": 1589.72,
                    "word": " of",
                    "probability": 0.904296875
                },
                {
                    "start": 1589.72,
                    "end": 1590.06,
                    "word": " differential",
                    "probability": 0.68896484375
                },
                {
                    "start": 1590.06,
                    "end": 1590.48,
                    "word": " privacy",
                    "probability": 0.8359375
                },
                {
                    "start": 1590.48,
                    "end": 1590.48,
                    "word": ",",
                    "probability": 0.98046875
                }
            ]
        },
        {
            "id": 249,
            "seek": 158939,
            "start": 1590.48,
            "end": 1593.76,
            "text": " we're trying to protect a user, not a browser.",
            "tokens": [
                51514,
                321,
                434,
                1382,
                281,
                2371,
                257,
                4195,
                11,
                406,
                257,
                11185,
                13,
                51614
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17571272060536502,
            "compression_ratio": 1.6125461254612545,
            "no_speech_prob": 0.0014095306396484375,
            "words": [
                {
                    "start": 1590.48,
                    "end": 1590.7,
                    "word": " we're",
                    "probability": 0.930908203125
                },
                {
                    "start": 1590.7,
                    "end": 1590.88,
                    "word": " trying",
                    "probability": 0.80224609375
                },
                {
                    "start": 1590.88,
                    "end": 1591.04,
                    "word": " to",
                    "probability": 0.904296875
                },
                {
                    "start": 1591.04,
                    "end": 1591.32,
                    "word": " protect",
                    "probability": 0.77392578125
                },
                {
                    "start": 1591.5,
                    "end": 1591.88,
                    "word": " a user",
                    "probability": 0.81298828125
                },
                {
                    "start": 1591.88,
                    "end": 1592.4,
                    "word": ",",
                    "probability": 0.927734375
                },
                {
                    "start": 1592.4,
                    "end": 1592.84,
                    "word": " not",
                    "probability": 0.88818359375
                },
                {
                    "start": 1592.98,
                    "end": 1593.28,
                    "word": " a browser",
                    "probability": 0.71142578125
                },
                {
                    "start": 1593.28,
                    "end": 1593.76,
                    "word": ".",
                    "probability": 0.9970703125
                }
            ]
        },
        {
            "id": 250,
            "seek": 161735,
            "start": 1593.76,
            "end": 1601.26,
            "text": " So if a user has used 10 different browsers and everybody else has just used one,",
            "tokens": [
                50364,
                407,
                498,
                257,
                4195,
                575,
                1143,
                1266,
                819,
                36069,
                293,
                2201,
                1646,
                575,
                445,
                1143,
                472,
                11,
                50664
            ],
            "temperature": 0.0,
            "avg_logprob": -0.21056548168971426,
            "compression_ratio": 1.6605504587155964,
            "no_speech_prob": 0.015655517578125,
            "words": [
                {
                    "start": 1593.76,
                    "end": 1594.1,
                    "word": " So",
                    "probability": 0.7021484375
                },
                {
                    "start": 1594.1,
                    "end": 1595.36,
                    "word": " if",
                    "probability": 0.6162109375
                },
                {
                    "start": 1595.92,
                    "end": 1596.28,
                    "word": " a user",
                    "probability": 0.8154296875
                },
                {
                    "start": 1596.28,
                    "end": 1596.5,
                    "word": " has",
                    "probability": 0.89306640625
                },
                {
                    "start": 1596.5,
                    "end": 1596.8,
                    "word": " used",
                    "probability": 0.82421875
                },
                {
                    "start": 1596.8,
                    "end": 1597.16,
                    "word": " 10",
                    "probability": 0.79248046875
                },
                {
                    "start": 1597.16,
                    "end": 1597.62,
                    "word": " different",
                    "probability": 0.70556640625
                },
                {
                    "start": 1597.62,
                    "end": 1598.08,
                    "word": " browsers",
                    "probability": 0.78466796875
                },
                {
                    "start": 1598.08,
                    "end": 1598.48,
                    "word": " and",
                    "probability": 0.7373046875
                },
                {
                    "start": 1598.48,
                    "end": 1598.82,
                    "word": " everybody",
                    "probability": 0.748046875
                },
                {
                    "start": 1598.82,
                    "end": 1599.14,
                    "word": " else",
                    "probability": 0.85498046875
                },
                {
                    "start": 1599.14,
                    "end": 1599.28,
                    "word": " has",
                    "probability": 0.8779296875
                },
                {
                    "start": 1599.28,
                    "end": 1599.54,
                    "word": " just",
                    "probability": 0.8837890625
                },
                {
                    "start": 1599.54,
                    "end": 1599.78,
                    "word": " used",
                    "probability": 0.8193359375
                },
                {
                    "start": 1599.78,
                    "end": 1600.16,
                    "word": " one",
                    "probability": 0.78955078125
                },
                {
                    "start": 1600.16,
                    "end": 1601.26,
                    "word": ",",
                    "probability": 0.91259765625
                }
            ]
        },
        {
            "id": 251,
            "seek": 161735,
            "start": 1601.26,
            "end": 1607.04,
            "text": " potentially you violated kind of the epsilon guarantees of your system,",
            "tokens": [
                50714,
                7263,
                291,
                33239,
                733,
                295,
                264,
                17889,
                32567,
                295,
                428,
                1185,
                11,
                50964
            ],
            "temperature": 0.0,
            "avg_logprob": -0.21056548168971426,
            "compression_ratio": 1.6605504587155964,
            "no_speech_prob": 0.015655517578125,
            "words": [
                {
                    "start": 1601.26,
                    "end": 1601.8,
                    "word": " potentially",
                    "probability": 0.67333984375
                },
                {
                    "start": 1601.8,
                    "end": 1602.98,
                    "word": " you",
                    "probability": 0.671875
                },
                {
                    "start": 1602.98,
                    "end": 1603.66,
                    "word": " violated",
                    "probability": 0.52099609375
                },
                {
                    "start": 1603.66,
                    "end": 1603.96,
                    "word": " kind",
                    "probability": 0.4521484375
                },
                {
                    "start": 1603.96,
                    "end": 1604.06,
                    "word": " of",
                    "probability": 0.90869140625
                },
                {
                    "start": 1604.06,
                    "end": 1604.3,
                    "word": " the",
                    "probability": 0.8173828125
                },
                {
                    "start": 1604.3,
                    "end": 1604.8,
                    "word": " epsilon",
                    "probability": 0.5439453125
                },
                {
                    "start": 1604.8,
                    "end": 1605.36,
                    "word": " guarantees",
                    "probability": 0.818359375
                },
                {
                    "start": 1605.36,
                    "end": 1606.5,
                    "word": " of",
                    "probability": 0.9013671875
                },
                {
                    "start": 1606.5,
                    "end": 1606.64,
                    "word": " your",
                    "probability": 0.8076171875
                },
                {
                    "start": 1606.64,
                    "end": 1606.98,
                    "word": " system",
                    "probability": 0.87353515625
                },
                {
                    "start": 1606.98,
                    "end": 1607.04,
                    "word": ",",
                    "probability": 0.17138671875
                }
            ]
        },
        {
            "id": 252,
            "seek": 161735,
            "start": 1607.04,
            "end": 1608.28,
            "text": " if you allow that.",
            "tokens": [
                51014,
                498,
                291,
                2089,
                300,
                13,
                51064
            ],
            "temperature": 0.0,
            "avg_logprob": -0.21056548168971426,
            "compression_ratio": 1.6605504587155964,
            "no_speech_prob": 0.015655517578125,
            "words": [
                {
                    "start": 1607.04,
                    "end": 1607.18,
                    "word": " if",
                    "probability": 0.90087890625
                },
                {
                    "start": 1607.18,
                    "end": 1607.28,
                    "word": " you",
                    "probability": 0.89697265625
                },
                {
                    "start": 1607.28,
                    "end": 1607.52,
                    "word": " allow",
                    "probability": 0.8798828125
                },
                {
                    "start": 1607.52,
                    "end": 1607.84,
                    "word": " that",
                    "probability": 0.88330078125
                },
                {
                    "start": 1607.84,
                    "end": 1608.28,
                    "word": ".",
                    "probability": 0.98828125
                }
            ]
        },
        {
            "id": 253,
            "seek": 161735,
            "start": 1608.28,
            "end": 1614.62,
            "text": " So the way that we deal with that is with reservoir sampling of a user's contributions",
            "tokens": [
                51114,
                407,
                264,
                636,
                300,
                321,
                2028,
                365,
                300,
                307,
                365,
                26316,
                21179,
                295,
                257,
                4195,
                311,
                15725,
                51364
            ],
            "temperature": 0.0,
            "avg_logprob": -0.21056548168971426,
            "compression_ratio": 1.6605504587155964,
            "no_speech_prob": 0.015655517578125,
            "words": [
                {
                    "start": 1608.28,
                    "end": 1608.4,
                    "word": " So",
                    "probability": 0.8935546875
                },
                {
                    "start": 1608.4,
                    "end": 1608.52,
                    "word": " the",
                    "probability": 0.78369140625
                },
                {
                    "start": 1608.52,
                    "end": 1608.64,
                    "word": " way",
                    "probability": 0.90478515625
                },
                {
                    "start": 1608.64,
                    "end": 1608.74,
                    "word": " that",
                    "probability": 0.8701171875
                },
                {
                    "start": 1608.74,
                    "end": 1608.82,
                    "word": " we",
                    "probability": 0.9013671875
                },
                {
                    "start": 1608.82,
                    "end": 1609.0,
                    "word": " deal",
                    "probability": 0.927734375
                },
                {
                    "start": 1609.0,
                    "end": 1609.16,
                    "word": " with",
                    "probability": 0.81689453125
                },
                {
                    "start": 1609.16,
                    "end": 1609.34,
                    "word": " that",
                    "probability": 0.88037109375
                },
                {
                    "start": 1609.34,
                    "end": 1609.5,
                    "word": " is",
                    "probability": 0.892578125
                },
                {
                    "start": 1609.5,
                    "end": 1610.02,
                    "word": " with",
                    "probability": 0.80859375
                },
                {
                    "start": 1610.02,
                    "end": 1611.58,
                    "word": " reservoir",
                    "probability": 0.61572265625
                },
                {
                    "start": 1611.58,
                    "end": 1612.06,
                    "word": " sampling",
                    "probability": 0.8505859375
                },
                {
                    "start": 1612.06,
                    "end": 1613.18,
                    "word": " of",
                    "probability": 0.904296875
                },
                {
                    "start": 1613.78,
                    "end": 1614.1,
                    "word": " a user's",
                    "probability": 0.880126953125
                },
                {
                    "start": 1614.1,
                    "end": 1614.62,
                    "word": " contributions",
                    "probability": 0.90625
                }
            ]
        },
        {
            "id": 254,
            "seek": 161735,
            "start": 1614.62,
            "end": 1618.16,
            "text": " and then limiting those contributions.",
            "tokens": [
                51414,
                293,
                550,
                22083,
                729,
                15725,
                13,
                51514
            ],
            "temperature": 0.0,
            "avg_logprob": -0.21056548168971426,
            "compression_ratio": 1.6605504587155964,
            "no_speech_prob": 0.015655517578125,
            "words": [
                {
                    "start": 1614.62,
                    "end": 1615.92,
                    "word": " and",
                    "probability": 0.496337890625
                },
                {
                    "start": 1615.92,
                    "end": 1616.12,
                    "word": " then",
                    "probability": 0.72314453125
                },
                {
                    "start": 1616.12,
                    "end": 1616.48,
                    "word": " limiting",
                    "probability": 0.8681640625
                },
                {
                    "start": 1616.48,
                    "end": 1616.78,
                    "word": " those",
                    "probability": 0.84521484375
                },
                {
                    "start": 1616.78,
                    "end": 1617.66,
                    "word": " contributions",
                    "probability": 0.92724609375
                },
                {
                    "start": 1617.66,
                    "end": 1618.16,
                    "word": ".",
                    "probability": 0.98046875
                }
            ]
        },
        {
            "id": 255,
            "seek": 161735,
            "start": 1618.16,
            "end": 1621.72,
            "text": " So a user might be able to contribute to 10 separate partitions",
            "tokens": [
                51564,
                407,
                257,
                4195,
                1062,
                312,
                1075,
                281,
                10586,
                281,
                1266,
                4994,
                644,
                2451,
                51714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.21056548168971426,
            "compression_ratio": 1.6605504587155964,
            "no_speech_prob": 0.015655517578125,
            "words": [
                {
                    "start": 1618.16,
                    "end": 1618.36,
                    "word": " So",
                    "probability": 0.8974609375
                },
                {
                    "start": 1618.78,
                    "end": 1618.92,
                    "word": " a user",
                    "probability": 0.79736328125
                },
                {
                    "start": 1618.92,
                    "end": 1619.1,
                    "word": " might",
                    "probability": 0.81494140625
                },
                {
                    "start": 1619.1,
                    "end": 1619.24,
                    "word": " be",
                    "probability": 0.90966796875
                },
                {
                    "start": 1619.24,
                    "end": 1619.36,
                    "word": " able",
                    "probability": 0.880859375
                },
                {
                    "start": 1619.36,
                    "end": 1619.8,
                    "word": " to",
                    "probability": 0.90380859375
                },
                {
                    "start": 1619.8,
                    "end": 1620.28,
                    "word": " contribute",
                    "probability": 0.85791015625
                },
                {
                    "start": 1620.28,
                    "end": 1620.58,
                    "word": " to",
                    "probability": 0.8994140625
                },
                {
                    "start": 1620.58,
                    "end": 1620.82,
                    "word": " 10",
                    "probability": 0.884765625
                },
                {
                    "start": 1620.82,
                    "end": 1621.2,
                    "word": " separate",
                    "probability": 0.8837890625
                },
                {
                    "start": 1621.2,
                    "end": 1621.72,
                    "word": " partitions",
                    "probability": 0.93994140625
                }
            ]
        },
        {
            "id": 256,
            "seek": 164735,
            "start": 1621.72,
            "end": 1624.38,
            "text": " or three or one or whatever seems appropriate.",
            "tokens": [
                50364,
                420,
                1045,
                420,
                472,
                420,
                2035,
                2544,
                6854,
                13,
                50464
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1727350972114353,
            "compression_ratio": 1.6703703703703703,
            "no_speech_prob": 0.015899658203125,
            "words": [
                {
                    "start": 1621.72,
                    "end": 1621.98,
                    "word": " or",
                    "probability": 0.358154296875
                },
                {
                    "start": 1621.98,
                    "end": 1622.28,
                    "word": " three",
                    "probability": 0.6083984375
                },
                {
                    "start": 1622.28,
                    "end": 1622.56,
                    "word": " or",
                    "probability": 0.60595703125
                },
                {
                    "start": 1622.56,
                    "end": 1622.92,
                    "word": " one",
                    "probability": 0.80859375
                },
                {
                    "start": 1622.92,
                    "end": 1623.04,
                    "word": " or",
                    "probability": 0.8076171875
                },
                {
                    "start": 1623.04,
                    "end": 1623.32,
                    "word": " whatever",
                    "probability": 0.82421875
                },
                {
                    "start": 1623.32,
                    "end": 1623.66,
                    "word": " seems",
                    "probability": 0.74267578125
                },
                {
                    "start": 1623.66,
                    "end": 1624.18,
                    "word": " appropriate",
                    "probability": 0.73681640625
                },
                {
                    "start": 1624.18,
                    "end": 1624.38,
                    "word": ".",
                    "probability": 0.76220703125
                }
            ]
        },
        {
            "id": 257,
            "seek": 164735,
            "start": 1624.38,
            "end": 1629.36,
            "text": " And again, it's another parameter that can be chosen by the user.",
            "tokens": [
                50514,
                400,
                797,
                11,
                309,
                311,
                1071,
                13075,
                300,
                393,
                312,
                8614,
                538,
                264,
                4195,
                13,
                50714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1727350972114353,
            "compression_ratio": 1.6703703703703703,
            "no_speech_prob": 0.015899658203125,
            "words": [
                {
                    "start": 1624.38,
                    "end": 1624.46,
                    "word": " And",
                    "probability": 0.85546875
                },
                {
                    "start": 1624.46,
                    "end": 1624.66,
                    "word": " again",
                    "probability": 0.857421875
                },
                {
                    "start": 1624.66,
                    "end": 1624.7,
                    "word": ",",
                    "probability": 0.8720703125
                },
                {
                    "start": 1624.7,
                    "end": 1624.98,
                    "word": " it's",
                    "probability": 0.948974609375
                },
                {
                    "start": 1624.98,
                    "end": 1625.22,
                    "word": " another",
                    "probability": 0.8251953125
                },
                {
                    "start": 1625.22,
                    "end": 1625.64,
                    "word": " parameter",
                    "probability": 0.8876953125
                },
                {
                    "start": 1625.64,
                    "end": 1627.26,
                    "word": " that",
                    "probability": 0.8134765625
                },
                {
                    "start": 1627.26,
                    "end": 1627.48,
                    "word": " can",
                    "probability": 0.89990234375
                },
                {
                    "start": 1627.48,
                    "end": 1627.66,
                    "word": " be",
                    "probability": 0.9013671875
                },
                {
                    "start": 1627.66,
                    "end": 1627.98,
                    "word": " chosen",
                    "probability": 0.91259765625
                },
                {
                    "start": 1627.98,
                    "end": 1628.32,
                    "word": " by",
                    "probability": 0.904296875
                },
                {
                    "start": 1628.32,
                    "end": 1628.5,
                    "word": " the",
                    "probability": 0.81396484375
                },
                {
                    "start": 1628.5,
                    "end": 1628.84,
                    "word": " user",
                    "probability": 0.794921875
                },
                {
                    "start": 1628.84,
                    "end": 1629.36,
                    "word": ".",
                    "probability": 0.287841796875
                }
            ]
        },
        {
            "id": 258,
            "seek": 164735,
            "start": 1629.48,
            "end": 1634.52,
            "text": " But you can also sort of choose one based on an appropriate epsilon and delta.",
            "tokens": [
                50764,
                583,
                291,
                393,
                611,
                1333,
                295,
                2826,
                472,
                2361,
                322,
                364,
                6854,
                17889,
                293,
                8289,
                13,
                50964
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1727350972114353,
            "compression_ratio": 1.6703703703703703,
            "no_speech_prob": 0.015899658203125,
            "words": [
                {
                    "start": 1629.48,
                    "end": 1629.72,
                    "word": " But",
                    "probability": 0.88818359375
                },
                {
                    "start": 1629.72,
                    "end": 1629.82,
                    "word": " you",
                    "probability": 0.88818359375
                },
                {
                    "start": 1629.82,
                    "end": 1629.96,
                    "word": " can",
                    "probability": 0.89892578125
                },
                {
                    "start": 1629.96,
                    "end": 1630.2,
                    "word": " also",
                    "probability": 0.8251953125
                },
                {
                    "start": 1630.2,
                    "end": 1630.48,
                    "word": " sort",
                    "probability": 0.78857421875
                },
                {
                    "start": 1630.48,
                    "end": 1630.76,
                    "word": " of",
                    "probability": 0.90380859375
                },
                {
                    "start": 1630.76,
                    "end": 1631.36,
                    "word": " choose",
                    "probability": 0.81396484375
                },
                {
                    "start": 1631.36,
                    "end": 1631.6,
                    "word": " one",
                    "probability": 0.8291015625
                },
                {
                    "start": 1631.6,
                    "end": 1631.9,
                    "word": " based",
                    "probability": 0.82275390625
                },
                {
                    "start": 1631.9,
                    "end": 1632.28,
                    "word": " on",
                    "probability": 0.90673828125
                },
                {
                    "start": 1632.28,
                    "end": 1632.52,
                    "word": " an",
                    "probability": 0.81591796875
                },
                {
                    "start": 1632.52,
                    "end": 1632.9,
                    "word": " appropriate",
                    "probability": 0.755859375
                },
                {
                    "start": 1632.9,
                    "end": 1633.3,
                    "word": " epsilon",
                    "probability": 0.67041015625
                },
                {
                    "start": 1633.3,
                    "end": 1633.72,
                    "word": " and",
                    "probability": 0.89208984375
                },
                {
                    "start": 1633.72,
                    "end": 1634.0,
                    "word": " delta",
                    "probability": 0.775390625
                },
                {
                    "start": 1634.0,
                    "end": 1634.52,
                    "word": ".",
                    "probability": 0.99072265625
                }
            ]
        },
        {
            "id": 259,
            "seek": 164735,
            "start": 1635.02,
            "end": 1640.16,
            "text": " So again, it's just about giving an engine the flexibility to represent",
            "tokens": [
                51014,
                407,
                797,
                11,
                309,
                311,
                445,
                466,
                2902,
                364,
                2848,
                264,
                12635,
                281,
                2906,
                51264
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1727350972114353,
            "compression_ratio": 1.6703703703703703,
            "no_speech_prob": 0.015899658203125,
            "words": [
                {
                    "start": 1635.02,
                    "end": 1635.16,
                    "word": " So",
                    "probability": 0.88720703125
                },
                {
                    "start": 1635.16,
                    "end": 1635.66,
                    "word": " again",
                    "probability": 0.67041015625
                },
                {
                    "start": 1635.66,
                    "end": 1635.68,
                    "word": ",",
                    "probability": 0.982421875
                },
                {
                    "start": 1635.68,
                    "end": 1635.9,
                    "word": " it's",
                    "probability": 0.952392578125
                },
                {
                    "start": 1635.9,
                    "end": 1636.1,
                    "word": " just",
                    "probability": 0.8837890625
                },
                {
                    "start": 1636.1,
                    "end": 1636.34,
                    "word": " about",
                    "probability": 0.80615234375
                },
                {
                    "start": 1636.34,
                    "end": 1636.78,
                    "word": " giving",
                    "probability": 0.8330078125
                },
                {
                    "start": 1636.78,
                    "end": 1638.58,
                    "word": " an",
                    "probability": 0.8857421875
                },
                {
                    "start": 1638.58,
                    "end": 1638.82,
                    "word": " engine",
                    "probability": 0.7392578125
                },
                {
                    "start": 1638.82,
                    "end": 1638.98,
                    "word": " the",
                    "probability": 0.72509765625
                },
                {
                    "start": 1638.98,
                    "end": 1639.3,
                    "word": " flexibility",
                    "probability": 0.9423828125
                },
                {
                    "start": 1639.3,
                    "end": 1639.68,
                    "word": " to",
                    "probability": 0.90673828125
                },
                {
                    "start": 1639.68,
                    "end": 1640.16,
                    "word": " represent",
                    "probability": 0.755859375
                }
            ]
        },
        {
            "id": 260,
            "seek": 164735,
            "start": 1640.16,
            "end": 1643.58,
            "text": " sort of both utility and privacy needs at the same time.",
            "tokens": [
                51314,
                1333,
                295,
                1293,
                14877,
                293,
                11427,
                2203,
                412,
                264,
                912,
                565,
                13,
                51414
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1727350972114353,
            "compression_ratio": 1.6703703703703703,
            "no_speech_prob": 0.015899658203125,
            "words": [
                {
                    "start": 1640.16,
                    "end": 1640.48,
                    "word": " sort",
                    "probability": 0.67822265625
                },
                {
                    "start": 1640.48,
                    "end": 1640.58,
                    "word": " of",
                    "probability": 0.9013671875
                },
                {
                    "start": 1640.58,
                    "end": 1640.78,
                    "word": " both",
                    "probability": 0.81005859375
                },
                {
                    "start": 1640.78,
                    "end": 1641.22,
                    "word": " utility",
                    "probability": 0.8427734375
                },
                {
                    "start": 1641.22,
                    "end": 1641.54,
                    "word": " and",
                    "probability": 0.89404296875
                },
                {
                    "start": 1641.54,
                    "end": 1641.86,
                    "word": " privacy",
                    "probability": 0.83056640625
                },
                {
                    "start": 1641.86,
                    "end": 1642.24,
                    "word": " needs",
                    "probability": 0.74853515625
                },
                {
                    "start": 1642.24,
                    "end": 1642.38,
                    "word": " at",
                    "probability": 0.90478515625
                },
                {
                    "start": 1642.38,
                    "end": 1642.5,
                    "word": " the",
                    "probability": 0.814453125
                },
                {
                    "start": 1642.5,
                    "end": 1642.68,
                    "word": " same",
                    "probability": 0.810546875
                },
                {
                    "start": 1642.68,
                    "end": 1643.06,
                    "word": " time",
                    "probability": 0.806640625
                },
                {
                    "start": 1643.06,
                    "end": 1643.58,
                    "word": ".",
                    "probability": 0.994140625
                }
            ]
        },
        {
            "id": 261,
            "seek": 164735,
            "start": 1644.04,
            "end": 1646.78,
            "text": " And this is basically the model in a nutshell.",
            "tokens": [
                51464,
                400,
                341,
                307,
                1936,
                264,
                2316,
                294,
                257,
                37711,
                13,
                51564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1727350972114353,
            "compression_ratio": 1.6703703703703703,
            "no_speech_prob": 0.015899658203125,
            "words": [
                {
                    "start": 1644.04,
                    "end": 1644.56,
                    "word": " And",
                    "probability": 0.8955078125
                },
                {
                    "start": 1644.56,
                    "end": 1644.74,
                    "word": " this",
                    "probability": 0.87841796875
                },
                {
                    "start": 1644.74,
                    "end": 1644.98,
                    "word": " is",
                    "probability": 0.90966796875
                },
                {
                    "start": 1644.98,
                    "end": 1645.58,
                    "word": " basically",
                    "probability": 0.89208984375
                },
                {
                    "start": 1645.58,
                    "end": 1645.82,
                    "word": " the",
                    "probability": 0.8134765625
                },
                {
                    "start": 1645.82,
                    "end": 1646.04,
                    "word": " model",
                    "probability": 0.82177734375
                },
                {
                    "start": 1646.04,
                    "end": 1646.2,
                    "word": " in",
                    "probability": 0.8974609375
                },
                {
                    "start": 1646.32,
                    "end": 1646.6,
                    "word": " a nutshell",
                    "probability": 0.72216796875
                },
                {
                    "start": 1646.6,
                    "end": 1646.78,
                    "word": ".",
                    "probability": 0.97119140625
                }
            ]
        },
        {
            "id": 262,
            "seek": 164735,
            "start": 1646.78,
            "end": 1650.96,
            "text": " Now, of course, if you were requiring users to write exactly the syntax every time,",
            "tokens": [
                51614,
                823,
                11,
                295,
                1164,
                11,
                498,
                291,
                645,
                24165,
                5022,
                281,
                2464,
                2293,
                264,
                28431,
                633,
                565,
                11,
                51814
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1727350972114353,
            "compression_ratio": 1.6703703703703703,
            "no_speech_prob": 0.015899658203125,
            "words": [
                {
                    "start": 1646.78,
                    "end": 1646.88,
                    "word": " Now",
                    "probability": 0.8818359375
                },
                {
                    "start": 1646.88,
                    "end": 1646.94,
                    "word": ",",
                    "probability": 0.7353515625
                },
                {
                    "start": 1646.94,
                    "end": 1647.04,
                    "word": " of",
                    "probability": 0.9052734375
                },
                {
                    "start": 1647.04,
                    "end": 1647.18,
                    "word": " course",
                    "probability": 0.86669921875
                },
                {
                    "start": 1647.18,
                    "end": 1647.18,
                    "word": ",",
                    "probability": 0.98046875
                },
                {
                    "start": 1647.18,
                    "end": 1647.36,
                    "word": " if",
                    "probability": 0.90087890625
                },
                {
                    "start": 1647.36,
                    "end": 1647.46,
                    "word": " you",
                    "probability": 0.89697265625
                },
                {
                    "start": 1647.46,
                    "end": 1647.62,
                    "word": " were",
                    "probability": 0.84716796875
                },
                {
                    "start": 1647.62,
                    "end": 1648.08,
                    "word": " requiring",
                    "probability": 0.935546875
                },
                {
                    "start": 1648.08,
                    "end": 1648.6,
                    "word": " users",
                    "probability": 0.82177734375
                },
                {
                    "start": 1648.6,
                    "end": 1648.86,
                    "word": " to",
                    "probability": 0.9052734375
                },
                {
                    "start": 1648.86,
                    "end": 1649.12,
                    "word": " write",
                    "probability": 0.80419921875
                },
                {
                    "start": 1649.12,
                    "end": 1649.62,
                    "word": " exactly",
                    "probability": 0.771484375
                },
                {
                    "start": 1649.62,
                    "end": 1649.86,
                    "word": " the",
                    "probability": 0.68017578125
                },
                {
                    "start": 1649.86,
                    "end": 1650.3,
                    "word": " syntax",
                    "probability": 0.92431640625
                },
                {
                    "start": 1650.3,
                    "end": 1650.58,
                    "word": " every",
                    "probability": 0.56982421875
                },
                {
                    "start": 1650.58,
                    "end": 1650.94,
                    "word": " time",
                    "probability": 0.8037109375
                },
                {
                    "start": 1650.94,
                    "end": 1650.96,
                    "word": ",",
                    "probability": 0.51806640625
                }
            ]
        },
        {
            "id": 263,
            "seek": 167393,
            "start": 1651.72,
            "end": 1652.12,
            "text": " that's not possible.",
            "tokens": [
                50364,
                300,
                311,
                406,
                1944,
                13,
                50414
            ],
            "temperature": 0.0,
            "avg_logprob": -0.26996528340710535,
            "compression_ratio": 1.6103896103896105,
            "no_speech_prob": 0.006488800048828125,
            "words": [
                {
                    "start": 1651.72,
                    "end": 1651.72,
                    "word": " that's",
                    "probability": 0.054622650146484375
                },
                {
                    "start": 1651.72,
                    "end": 1651.72,
                    "word": " not",
                    "probability": 0.05718994140625
                },
                {
                    "start": 1651.72,
                    "end": 1651.72,
                    "word": " possible",
                    "probability": 0.005870819091796875
                },
                {
                    "start": 1651.72,
                    "end": 1652.12,
                    "word": ".",
                    "probability": 0.56689453125
                }
            ]
        },
        {
            "id": 264,
            "seek": 167393,
            "start": 1652.12,
            "end": 1658.24,
            "text": " And so the new syntax is introduced with anonymization is basically just a clever rewriter",
            "tokens": [
                50464,
                400,
                370,
                264,
                777,
                28431,
                307,
                7268,
                365,
                37293,
                2144,
                307,
                1936,
                445,
                257,
                13494,
                319,
                23681,
                50714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.26996528340710535,
            "compression_ratio": 1.6103896103896105,
            "no_speech_prob": 0.006488800048828125,
            "words": [
                {
                    "start": 1652.12,
                    "end": 1652.42,
                    "word": " And",
                    "probability": 0.724609375
                },
                {
                    "start": 1652.42,
                    "end": 1652.88,
                    "word": " so",
                    "probability": 0.88916015625
                },
                {
                    "start": 1652.88,
                    "end": 1653.22,
                    "word": " the",
                    "probability": 0.6533203125
                },
                {
                    "start": 1653.22,
                    "end": 1653.54,
                    "word": " new",
                    "probability": 0.798828125
                },
                {
                    "start": 1653.54,
                    "end": 1654.02,
                    "word": " syntax",
                    "probability": 0.9169921875
                },
                {
                    "start": 1654.02,
                    "end": 1654.2,
                    "word": " is",
                    "probability": 0.87353515625
                },
                {
                    "start": 1654.2,
                    "end": 1654.72,
                    "word": " introduced",
                    "probability": 0.7548828125
                },
                {
                    "start": 1654.72,
                    "end": 1655.3,
                    "word": " with",
                    "probability": 0.12237548828125
                },
                {
                    "start": 1655.3,
                    "end": 1656.1,
                    "word": " anonymization",
                    "probability": 0.83642578125
                },
                {
                    "start": 1656.1,
                    "end": 1656.76,
                    "word": " is",
                    "probability": 0.67822265625
                },
                {
                    "start": 1656.76,
                    "end": 1657.12,
                    "word": " basically",
                    "probability": 0.857421875
                },
                {
                    "start": 1657.12,
                    "end": 1657.4,
                    "word": " just",
                    "probability": 0.88134765625
                },
                {
                    "start": 1657.54,
                    "end": 1657.78,
                    "word": " a clever",
                    "probability": 0.88671875
                },
                {
                    "start": 1657.78,
                    "end": 1658.24,
                    "word": " rewriter",
                    "probability": 0.734375
                }
            ]
        },
        {
            "id": 265,
            "seek": 167393,
            "start": 1658.24,
            "end": 1662.7,
            "text": " that is doing all of that in the background to the AST of the SQL",
            "tokens": [
                50764,
                300,
                307,
                884,
                439,
                295,
                300,
                294,
                264,
                3678,
                281,
                264,
                316,
                6840,
                295,
                264,
                19200,
                50964
            ],
            "temperature": 0.0,
            "avg_logprob": -0.26996528340710535,
            "compression_ratio": 1.6103896103896105,
            "no_speech_prob": 0.006488800048828125,
            "words": [
                {
                    "start": 1658.24,
                    "end": 1658.82,
                    "word": " that",
                    "probability": 0.82763671875
                },
                {
                    "start": 1658.82,
                    "end": 1658.94,
                    "word": " is",
                    "probability": 0.89990234375
                },
                {
                    "start": 1658.94,
                    "end": 1659.16,
                    "word": " doing",
                    "probability": 0.90673828125
                },
                {
                    "start": 1659.16,
                    "end": 1659.36,
                    "word": " all",
                    "probability": 0.89453125
                },
                {
                    "start": 1659.36,
                    "end": 1659.46,
                    "word": " of",
                    "probability": 0.88427734375
                },
                {
                    "start": 1659.46,
                    "end": 1659.62,
                    "word": " that",
                    "probability": 0.87548828125
                },
                {
                    "start": 1659.62,
                    "end": 1659.72,
                    "word": " in",
                    "probability": 0.8349609375
                },
                {
                    "start": 1659.72,
                    "end": 1659.84,
                    "word": " the",
                    "probability": 0.75390625
                },
                {
                    "start": 1659.84,
                    "end": 1660.5,
                    "word": " background",
                    "probability": 0.68505859375
                },
                {
                    "start": 1660.5,
                    "end": 1661.08,
                    "word": " to",
                    "probability": 0.890625
                },
                {
                    "start": 1661.08,
                    "end": 1661.28,
                    "word": " the",
                    "probability": 0.81689453125
                },
                {
                    "start": 1661.28,
                    "end": 1661.72,
                    "word": " AST",
                    "probability": 0.8857421875
                },
                {
                    "start": 1661.72,
                    "end": 1662.26,
                    "word": " of",
                    "probability": 0.89892578125
                },
                {
                    "start": 1662.26,
                    "end": 1662.4,
                    "word": " the",
                    "probability": 0.8125
                },
                {
                    "start": 1662.4,
                    "end": 1662.7,
                    "word": " SQL",
                    "probability": 0.783203125
                }
            ]
        },
        {
            "id": 266,
            "seek": 167393,
            "start": 1663.28,
            "end": 1668.54,
            "text": " and allowing people to write basically SQL that they're familiar with.",
            "tokens": [
                51014,
                293,
                8293,
                561,
                281,
                2464,
                1936,
                19200,
                300,
                436,
                434,
                4963,
                365,
                13,
                51214
            ],
            "temperature": 0.0,
            "avg_logprob": -0.26996528340710535,
            "compression_ratio": 1.6103896103896105,
            "no_speech_prob": 0.006488800048828125,
            "words": [
                {
                    "start": 1663.28,
                    "end": 1663.86,
                    "word": " and",
                    "probability": 0.5986328125
                },
                {
                    "start": 1663.86,
                    "end": 1664.12,
                    "word": " allowing",
                    "probability": 0.791015625
                },
                {
                    "start": 1664.12,
                    "end": 1664.34,
                    "word": " people",
                    "probability": 0.92822265625
                },
                {
                    "start": 1664.34,
                    "end": 1664.5,
                    "word": " to",
                    "probability": 0.90234375
                },
                {
                    "start": 1664.5,
                    "end": 1664.74,
                    "word": " write",
                    "probability": 0.82666015625
                },
                {
                    "start": 1664.74,
                    "end": 1665.54,
                    "word": " basically",
                    "probability": 0.7685546875
                },
                {
                    "start": 1665.54,
                    "end": 1667.16,
                    "word": " SQL",
                    "probability": 0.859375
                },
                {
                    "start": 1667.16,
                    "end": 1667.34,
                    "word": " that",
                    "probability": 0.87939453125
                },
                {
                    "start": 1667.34,
                    "end": 1667.52,
                    "word": " they're",
                    "probability": 0.825439453125
                },
                {
                    "start": 1667.52,
                    "end": 1667.82,
                    "word": " familiar",
                    "probability": 0.775390625
                },
                {
                    "start": 1667.82,
                    "end": 1668.2,
                    "word": " with",
                    "probability": 0.8115234375
                },
                {
                    "start": 1668.2,
                    "end": 1668.54,
                    "word": ".",
                    "probability": 0.9521484375
                }
            ]
        },
        {
            "id": 267,
            "seek": 167393,
            "start": 1668.54,
            "end": 1672.58,
            "text": " You see a few more keywords in here, obviously, the privacy parameters,",
            "tokens": [
                51264,
                509,
                536,
                257,
                1326,
                544,
                21009,
                294,
                510,
                11,
                2745,
                11,
                264,
                11427,
                9834,
                11,
                51414
            ],
            "temperature": 0.0,
            "avg_logprob": -0.26996528340710535,
            "compression_ratio": 1.6103896103896105,
            "no_speech_prob": 0.006488800048828125,
            "words": [
                {
                    "start": 1668.54,
                    "end": 1668.62,
                    "word": " You",
                    "probability": 0.89013671875
                },
                {
                    "start": 1668.62,
                    "end": 1668.76,
                    "word": " see",
                    "probability": 0.81494140625
                },
                {
                    "start": 1668.88,
                    "end": 1668.96,
                    "word": " a few",
                    "probability": 0.8056640625
                },
                {
                    "start": 1668.96,
                    "end": 1669.18,
                    "word": " more",
                    "probability": 0.89892578125
                },
                {
                    "start": 1669.18,
                    "end": 1669.56,
                    "word": " keywords",
                    "probability": 0.59423828125
                },
                {
                    "start": 1669.56,
                    "end": 1669.8,
                    "word": " in",
                    "probability": 0.8974609375
                },
                {
                    "start": 1669.8,
                    "end": 1670.0,
                    "word": " here",
                    "probability": 0.81640625
                },
                {
                    "start": 1670.0,
                    "end": 1670.08,
                    "word": ",",
                    "probability": 0.6318359375
                },
                {
                    "start": 1670.08,
                    "end": 1670.38,
                    "word": " obviously",
                    "probability": 0.7958984375
                },
                {
                    "start": 1670.38,
                    "end": 1670.64,
                    "word": ",",
                    "probability": 0.26123046875
                },
                {
                    "start": 1670.64,
                    "end": 1670.74,
                    "word": " the",
                    "probability": 0.77490234375
                },
                {
                    "start": 1670.74,
                    "end": 1671.1,
                    "word": " privacy",
                    "probability": 0.80078125
                },
                {
                    "start": 1671.1,
                    "end": 1671.7,
                    "word": " parameters",
                    "probability": 0.88671875
                },
                {
                    "start": 1671.7,
                    "end": 1672.58,
                    "word": ",",
                    "probability": 0.93017578125
                }
            ]
        },
        {
            "id": 268,
            "seek": 167393,
            "start": 1672.58,
            "end": 1678.3,
            "text": " the anonymization keyword in the aggregation calls.",
            "tokens": [
                51464,
                264,
                37293,
                2144,
                20428,
                294,
                264,
                16743,
                399,
                5498,
                13,
                51714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.26996528340710535,
            "compression_ratio": 1.6103896103896105,
            "no_speech_prob": 0.006488800048828125,
            "words": [
                {
                    "start": 1672.58,
                    "end": 1672.66,
                    "word": " the",
                    "probability": 0.80810546875
                },
                {
                    "start": 1672.66,
                    "end": 1673.26,
                    "word": " anonymization",
                    "probability": 0.885009765625
                },
                {
                    "start": 1673.26,
                    "end": 1673.64,
                    "word": " keyword",
                    "probability": 0.740234375
                },
                {
                    "start": 1673.64,
                    "end": 1673.84,
                    "word": " in",
                    "probability": 0.473876953125
                },
                {
                    "start": 1673.84,
                    "end": 1673.96,
                    "word": " the",
                    "probability": 0.81298828125
                },
                {
                    "start": 1673.96,
                    "end": 1674.62,
                    "word": " aggregation",
                    "probability": 0.86572265625
                },
                {
                    "start": 1674.62,
                    "end": 1678.02,
                    "word": " calls",
                    "probability": 0.6474609375
                },
                {
                    "start": 1678.02,
                    "end": 1678.3,
                    "word": ".",
                    "probability": 0.6767578125
                }
            ]
        },
        {
            "id": 269,
            "seek": 170381,
            "start": 1679.54,
            "end": 1684.5,
            "text": " And of course, sort of importantly, it's really critical to differential privacy",
            "tokens": [
                50364,
                400,
                295,
                1164,
                11,
                1333,
                295,
                8906,
                11,
                309,
                311,
                534,
                4924,
                281,
                15756,
                11427,
                50614
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19489247824556083,
            "compression_ratio": 1.7452830188679245,
            "no_speech_prob": 0.01204681396484375,
            "words": [
                {
                    "start": 1679.54,
                    "end": 1680.18,
                    "word": " And",
                    "probability": 0.30419921875
                },
                {
                    "start": 1680.18,
                    "end": 1680.36,
                    "word": " of",
                    "probability": 0.72802734375
                },
                {
                    "start": 1680.36,
                    "end": 1680.58,
                    "word": " course",
                    "probability": 0.8623046875
                },
                {
                    "start": 1680.58,
                    "end": 1681.06,
                    "word": ",",
                    "probability": 0.86279296875
                },
                {
                    "start": 1681.06,
                    "end": 1681.24,
                    "word": " sort",
                    "probability": 0.56005859375
                },
                {
                    "start": 1681.24,
                    "end": 1681.4,
                    "word": " of",
                    "probability": 0.9033203125
                },
                {
                    "start": 1681.4,
                    "end": 1682.16,
                    "word": " importantly",
                    "probability": 0.59521484375
                },
                {
                    "start": 1682.16,
                    "end": 1682.48,
                    "word": ",",
                    "probability": 0.84375
                },
                {
                    "start": 1682.48,
                    "end": 1682.6,
                    "word": " it's",
                    "probability": 0.9404296875
                },
                {
                    "start": 1682.6,
                    "end": 1683.14,
                    "word": " really",
                    "probability": 0.7763671875
                },
                {
                    "start": 1683.14,
                    "end": 1683.54,
                    "word": " critical",
                    "probability": 0.84375
                },
                {
                    "start": 1683.54,
                    "end": 1683.7,
                    "word": " to",
                    "probability": 0.7822265625
                },
                {
                    "start": 1683.7,
                    "end": 1684.02,
                    "word": " differential",
                    "probability": 0.64404296875
                },
                {
                    "start": 1684.02,
                    "end": 1684.5,
                    "word": " privacy",
                    "probability": 0.85546875
                }
            ]
        },
        {
            "id": 270,
            "seek": 170381,
            "start": 1684.5,
            "end": 1687.96,
            "text": " to be able to bound what a single user is able to contribute.",
            "tokens": [
                50664,
                281,
                312,
                1075,
                281,
                5472,
                437,
                257,
                2167,
                4195,
                307,
                1075,
                281,
                10586,
                13,
                50814
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19489247824556083,
            "compression_ratio": 1.7452830188679245,
            "no_speech_prob": 0.01204681396484375,
            "words": [
                {
                    "start": 1684.5,
                    "end": 1684.78,
                    "word": " to",
                    "probability": 0.875
                },
                {
                    "start": 1684.78,
                    "end": 1684.92,
                    "word": " be",
                    "probability": 0.884765625
                },
                {
                    "start": 1684.92,
                    "end": 1685.04,
                    "word": " able",
                    "probability": 0.8681640625
                },
                {
                    "start": 1685.04,
                    "end": 1685.2,
                    "word": " to",
                    "probability": 0.90283203125
                },
                {
                    "start": 1685.2,
                    "end": 1685.56,
                    "word": " bound",
                    "probability": 0.80419921875
                },
                {
                    "start": 1685.56,
                    "end": 1685.76,
                    "word": " what",
                    "probability": 0.8857421875
                },
                {
                    "start": 1685.88,
                    "end": 1686.1,
                    "word": " a single",
                    "probability": 0.888671875
                },
                {
                    "start": 1686.1,
                    "end": 1686.4,
                    "word": " user",
                    "probability": 0.81005859375
                },
                {
                    "start": 1686.4,
                    "end": 1686.6,
                    "word": " is",
                    "probability": 0.90087890625
                },
                {
                    "start": 1686.6,
                    "end": 1686.78,
                    "word": " able",
                    "probability": 0.890625
                },
                {
                    "start": 1686.78,
                    "end": 1686.94,
                    "word": " to",
                    "probability": 0.90380859375
                },
                {
                    "start": 1686.94,
                    "end": 1687.32,
                    "word": " contribute",
                    "probability": 0.87158203125
                },
                {
                    "start": 1687.32,
                    "end": 1687.96,
                    "word": ".",
                    "probability": 0.97265625
                }
            ]
        },
        {
            "id": 271,
            "seek": 170381,
            "start": 1688.42,
            "end": 1694.72,
            "text": " So, you know, from the perspective of a particular partition,",
            "tokens": [
                50864,
                407,
                11,
                291,
                458,
                11,
                490,
                264,
                4585,
                295,
                257,
                1729,
                24808,
                11,
                51114
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19489247824556083,
            "compression_ratio": 1.7452830188679245,
            "no_speech_prob": 0.01204681396484375,
            "words": [
                {
                    "start": 1688.42,
                    "end": 1689.06,
                    "word": " So",
                    "probability": 0.87841796875
                },
                {
                    "start": 1689.06,
                    "end": 1689.52,
                    "word": ",",
                    "probability": 0.66455078125
                },
                {
                    "start": 1689.52,
                    "end": 1689.8,
                    "word": " you",
                    "probability": 0.61767578125
                },
                {
                    "start": 1689.8,
                    "end": 1689.92,
                    "word": " know",
                    "probability": 0.81591796875
                },
                {
                    "start": 1689.92,
                    "end": 1689.98,
                    "word": ",",
                    "probability": 0.96533203125
                },
                {
                    "start": 1689.98,
                    "end": 1690.22,
                    "word": " from",
                    "probability": 0.81689453125
                },
                {
                    "start": 1690.22,
                    "end": 1690.5,
                    "word": " the",
                    "probability": 0.8203125
                },
                {
                    "start": 1690.5,
                    "end": 1691.04,
                    "word": " perspective",
                    "probability": 0.7666015625
                },
                {
                    "start": 1691.04,
                    "end": 1692.16,
                    "word": " of",
                    "probability": 0.90234375
                },
                {
                    "start": 1693.06,
                    "end": 1693.4,
                    "word": " a particular",
                    "probability": 0.80712890625
                },
                {
                    "start": 1693.4,
                    "end": 1693.98,
                    "word": " partition",
                    "probability": 0.75048828125
                },
                {
                    "start": 1693.98,
                    "end": 1694.72,
                    "word": ",",
                    "probability": 0.89404296875
                }
            ]
        },
        {
            "id": 272,
            "seek": 170381,
            "start": 1694.72,
            "end": 1701.24,
            "text": " perhaps, you know, you only want a user to be able to contribute a single visit to a website.",
            "tokens": [
                51164,
                4317,
                11,
                291,
                458,
                11,
                291,
                787,
                528,
                257,
                4195,
                281,
                312,
                1075,
                281,
                10586,
                257,
                2167,
                3441,
                281,
                257,
                3144,
                13,
                51464
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19489247824556083,
            "compression_ratio": 1.7452830188679245,
            "no_speech_prob": 0.01204681396484375,
            "words": [
                {
                    "start": 1694.72,
                    "end": 1695.52,
                    "word": " perhaps",
                    "probability": 0.81396484375
                },
                {
                    "start": 1695.52,
                    "end": 1696.06,
                    "word": ",",
                    "probability": 0.67236328125
                },
                {
                    "start": 1696.06,
                    "end": 1696.38,
                    "word": " you",
                    "probability": 0.89453125
                },
                {
                    "start": 1696.38,
                    "end": 1696.48,
                    "word": " know",
                    "probability": 0.81005859375
                },
                {
                    "start": 1696.48,
                    "end": 1696.54,
                    "word": ",",
                    "probability": 0.982421875
                },
                {
                    "start": 1696.54,
                    "end": 1696.62,
                    "word": " you",
                    "probability": 0.89404296875
                },
                {
                    "start": 1696.62,
                    "end": 1696.86,
                    "word": " only",
                    "probability": 0.8623046875
                },
                {
                    "start": 1696.86,
                    "end": 1697.04,
                    "word": " want",
                    "probability": 0.79248046875
                },
                {
                    "start": 1697.38,
                    "end": 1697.62,
                    "word": " a user",
                    "probability": 0.814453125
                },
                {
                    "start": 1697.62,
                    "end": 1697.86,
                    "word": " to",
                    "probability": 0.90625
                },
                {
                    "start": 1697.86,
                    "end": 1698.02,
                    "word": " be",
                    "probability": 0.91455078125
                },
                {
                    "start": 1698.02,
                    "end": 1698.14,
                    "word": " able",
                    "probability": 0.8916015625
                },
                {
                    "start": 1698.14,
                    "end": 1698.28,
                    "word": " to",
                    "probability": 0.90625
                },
                {
                    "start": 1698.28,
                    "end": 1698.6,
                    "word": " contribute",
                    "probability": 0.87646484375
                },
                {
                    "start": 1698.96,
                    "end": 1699.36,
                    "word": " a single",
                    "probability": 0.88427734375
                },
                {
                    "start": 1699.36,
                    "end": 1699.84,
                    "word": " visit",
                    "probability": 0.8408203125
                },
                {
                    "start": 1699.84,
                    "end": 1700.1,
                    "word": " to",
                    "probability": 0.90087890625
                },
                {
                    "start": 1700.26,
                    "end": 1700.64,
                    "word": " a website",
                    "probability": 0.8271484375
                },
                {
                    "start": 1700.64,
                    "end": 1701.24,
                    "word": ".",
                    "probability": 0.650390625
                }
            ]
        },
        {
            "id": 273,
            "seek": 170381,
            "start": 1701.5,
            "end": 1708.18,
            "text": " But perhaps for a salary, you'd want to allow them to contribute up to,",
            "tokens": [
                51514,
                583,
                4317,
                337,
                257,
                15360,
                11,
                291,
                1116,
                528,
                281,
                2089,
                552,
                281,
                10586,
                493,
                281,
                11,
                51764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19489247824556083,
            "compression_ratio": 1.7452830188679245,
            "no_speech_prob": 0.01204681396484375,
            "words": [
                {
                    "start": 1701.5,
                    "end": 1702.14,
                    "word": " But",
                    "probability": 0.89306640625
                },
                {
                    "start": 1702.14,
                    "end": 1702.5,
                    "word": " perhaps",
                    "probability": 0.8193359375
                },
                {
                    "start": 1702.5,
                    "end": 1703.0,
                    "word": " for",
                    "probability": 0.7138671875
                },
                {
                    "start": 1703.16,
                    "end": 1703.52,
                    "word": " a salary",
                    "probability": 0.83984375
                },
                {
                    "start": 1703.52,
                    "end": 1703.86,
                    "word": ",",
                    "probability": 0.8076171875
                },
                {
                    "start": 1703.86,
                    "end": 1704.3,
                    "word": " you'd",
                    "probability": 0.906982421875
                },
                {
                    "start": 1704.3,
                    "end": 1704.48,
                    "word": " want",
                    "probability": 0.75048828125
                },
                {
                    "start": 1704.48,
                    "end": 1704.84,
                    "word": " to",
                    "probability": 0.89892578125
                },
                {
                    "start": 1704.84,
                    "end": 1705.82,
                    "word": " allow",
                    "probability": 0.8759765625
                },
                {
                    "start": 1705.82,
                    "end": 1706.06,
                    "word": " them",
                    "probability": 0.73486328125
                },
                {
                    "start": 1706.06,
                    "end": 1706.34,
                    "word": " to",
                    "probability": 0.90380859375
                },
                {
                    "start": 1706.34,
                    "end": 1706.72,
                    "word": " contribute",
                    "probability": 0.8681640625
                },
                {
                    "start": 1706.72,
                    "end": 1707.06,
                    "word": " up",
                    "probability": 0.8955078125
                },
                {
                    "start": 1707.06,
                    "end": 1707.42,
                    "word": " to",
                    "probability": 0.8984375
                },
                {
                    "start": 1707.42,
                    "end": 1708.18,
                    "word": ",",
                    "probability": 0.7509765625
                }
            ]
        },
        {
            "id": 274,
            "seek": 173381,
            "start": 1708.18,
            "end": 1711.02,
            "text": " you know, $1 million, $1 million, whatever seems reasonable there.",
            "tokens": [
                50364,
                291,
                458,
                11,
                1848,
                16,
                2459,
                11,
                1848,
                16,
                2459,
                11,
                2035,
                2544,
                10585,
                456,
                13,
                50464
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17625899516421256,
            "compression_ratio": 1.7491961414790997,
            "no_speech_prob": 0.042083740234375,
            "words": [
                {
                    "start": 1708.18,
                    "end": 1708.54,
                    "word": " you",
                    "probability": 0.00270843505859375
                },
                {
                    "start": 1708.54,
                    "end": 1708.54,
                    "word": " know",
                    "probability": 0.1307373046875
                },
                {
                    "start": 1708.54,
                    "end": 1708.54,
                    "word": ",",
                    "probability": 0.76220703125
                },
                {
                    "start": 1708.54,
                    "end": 1708.54,
                    "word": " $1",
                    "probability": 0.4278564453125
                },
                {
                    "start": 1708.54,
                    "end": 1708.54,
                    "word": " million",
                    "probability": 0.57861328125
                },
                {
                    "start": 1708.54,
                    "end": 1708.62,
                    "word": ",",
                    "probability": 0.84814453125
                },
                {
                    "start": 1708.62,
                    "end": 1708.72,
                    "word": " $1",
                    "probability": 0.5869140625
                },
                {
                    "start": 1708.72,
                    "end": 1708.94,
                    "word": " million",
                    "probability": 0.6982421875
                },
                {
                    "start": 1708.94,
                    "end": 1709.08,
                    "word": ",",
                    "probability": 0.85107421875
                },
                {
                    "start": 1709.08,
                    "end": 1709.14,
                    "word": " whatever",
                    "probability": 0.650390625
                },
                {
                    "start": 1709.14,
                    "end": 1709.62,
                    "word": " seems",
                    "probability": 0.6796875
                },
                {
                    "start": 1709.62,
                    "end": 1710.26,
                    "word": " reasonable",
                    "probability": 0.7666015625
                },
                {
                    "start": 1710.26,
                    "end": 1710.58,
                    "word": " there",
                    "probability": 0.8427734375
                },
                {
                    "start": 1710.58,
                    "end": 1711.02,
                    "word": ".",
                    "probability": 0.8271484375
                }
            ]
        },
        {
            "id": 275,
            "seek": 173381,
            "start": 1711.28,
            "end": 1714.92,
            "text": " Perhaps if you were contributing an average age or something,",
            "tokens": [
                50514,
                10517,
                498,
                291,
                645,
                19270,
                364,
                4274,
                3205,
                420,
                746,
                11,
                50664
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17625899516421256,
            "compression_ratio": 1.7491961414790997,
            "no_speech_prob": 0.042083740234375,
            "words": [
                {
                    "start": 1711.28,
                    "end": 1711.64,
                    "word": " Perhaps",
                    "probability": 0.81494140625
                },
                {
                    "start": 1711.64,
                    "end": 1711.82,
                    "word": " if",
                    "probability": 0.86279296875
                },
                {
                    "start": 1711.82,
                    "end": 1712.1,
                    "word": " you",
                    "probability": 0.88330078125
                },
                {
                    "start": 1712.1,
                    "end": 1712.24,
                    "word": " were",
                    "probability": 0.79052734375
                },
                {
                    "start": 1712.24,
                    "end": 1712.72,
                    "word": " contributing",
                    "probability": 0.8544921875
                },
                {
                    "start": 1712.72,
                    "end": 1713.24,
                    "word": " an",
                    "probability": 0.859375
                },
                {
                    "start": 1713.24,
                    "end": 1713.88,
                    "word": " average",
                    "probability": 0.677734375
                },
                {
                    "start": 1713.88,
                    "end": 1714.34,
                    "word": " age",
                    "probability": 0.8994140625
                },
                {
                    "start": 1714.34,
                    "end": 1714.48,
                    "word": " or",
                    "probability": 0.89306640625
                },
                {
                    "start": 1714.48,
                    "end": 1714.82,
                    "word": " something",
                    "probability": 0.73486328125
                },
                {
                    "start": 1714.82,
                    "end": 1714.92,
                    "word": ",",
                    "probability": 0.849609375
                }
            ]
        },
        {
            "id": 276,
            "seek": 173381,
            "start": 1714.92,
            "end": 1717.7,
            "text": " you'd want to bound people between 0 and 120.",
            "tokens": [
                50714,
                291,
                1116,
                528,
                281,
                5472,
                561,
                1296,
                1958,
                293,
                10411,
                13,
                50814
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17625899516421256,
            "compression_ratio": 1.7491961414790997,
            "no_speech_prob": 0.042083740234375,
            "words": [
                {
                    "start": 1714.92,
                    "end": 1715.08,
                    "word": " you'd",
                    "probability": 0.9375
                },
                {
                    "start": 1715.08,
                    "end": 1715.22,
                    "word": " want",
                    "probability": 0.7021484375
                },
                {
                    "start": 1715.22,
                    "end": 1715.36,
                    "word": " to",
                    "probability": 0.8994140625
                },
                {
                    "start": 1715.36,
                    "end": 1715.6,
                    "word": " bound",
                    "probability": 0.84521484375
                },
                {
                    "start": 1715.6,
                    "end": 1715.94,
                    "word": " people",
                    "probability": 0.9248046875
                },
                {
                    "start": 1715.94,
                    "end": 1716.24,
                    "word": " between",
                    "probability": 0.76611328125
                },
                {
                    "start": 1716.24,
                    "end": 1716.54,
                    "word": " 0",
                    "probability": 0.2705078125
                },
                {
                    "start": 1716.54,
                    "end": 1716.7,
                    "word": " and",
                    "probability": 0.87158203125
                },
                {
                    "start": 1716.7,
                    "end": 1717.26,
                    "word": " 120",
                    "probability": 0.900390625
                },
                {
                    "start": 1717.26,
                    "end": 1717.7,
                    "word": ".",
                    "probability": 0.94482421875
                }
            ]
        },
        {
            "id": 277,
            "seek": 173381,
            "start": 1718.32,
            "end": 1722.74,
            "text": " The more that you can definitely say about aggregation bounds,",
            "tokens": [
                50864,
                440,
                544,
                300,
                291,
                393,
                2138,
                584,
                466,
                16743,
                399,
                29905,
                11,
                51064
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17625899516421256,
            "compression_ratio": 1.7491961414790997,
            "no_speech_prob": 0.042083740234375,
            "words": [
                {
                    "start": 1718.32,
                    "end": 1718.76,
                    "word": " The",
                    "probability": 0.8046875
                },
                {
                    "start": 1718.76,
                    "end": 1719.04,
                    "word": " more",
                    "probability": 0.87841796875
                },
                {
                    "start": 1719.04,
                    "end": 1719.2,
                    "word": " that",
                    "probability": 0.8623046875
                },
                {
                    "start": 1719.2,
                    "end": 1719.32,
                    "word": " you",
                    "probability": 0.8935546875
                },
                {
                    "start": 1719.32,
                    "end": 1719.54,
                    "word": " can",
                    "probability": 0.89501953125
                },
                {
                    "start": 1719.54,
                    "end": 1720.06,
                    "word": " definitely",
                    "probability": 0.7841796875
                },
                {
                    "start": 1720.06,
                    "end": 1720.54,
                    "word": " say",
                    "probability": 0.89697265625
                },
                {
                    "start": 1720.54,
                    "end": 1721.02,
                    "word": " about",
                    "probability": 0.8203125
                },
                {
                    "start": 1721.02,
                    "end": 1721.86,
                    "word": " aggregation",
                    "probability": 0.8525390625
                },
                {
                    "start": 1721.86,
                    "end": 1722.54,
                    "word": " bounds",
                    "probability": 0.68408203125
                },
                {
                    "start": 1722.54,
                    "end": 1722.74,
                    "word": ",",
                    "probability": 0.94140625
                }
            ]
        },
        {
            "id": 278,
            "seek": 173381,
            "start": 1722.74,
            "end": 1724.56,
            "text": " the more accuracy you can do.",
            "tokens": [
                51114,
                264,
                544,
                14170,
                291,
                393,
                360,
                13,
                51164
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17625899516421256,
            "compression_ratio": 1.7491961414790997,
            "no_speech_prob": 0.042083740234375,
            "words": [
                {
                    "start": 1722.74,
                    "end": 1723.18,
                    "word": " the",
                    "probability": 0.80810546875
                },
                {
                    "start": 1723.18,
                    "end": 1723.36,
                    "word": " more",
                    "probability": 0.87646484375
                },
                {
                    "start": 1723.36,
                    "end": 1723.76,
                    "word": " accuracy",
                    "probability": 0.76513671875
                },
                {
                    "start": 1723.76,
                    "end": 1723.96,
                    "word": " you",
                    "probability": 0.88818359375
                },
                {
                    "start": 1723.96,
                    "end": 1724.1,
                    "word": " can",
                    "probability": 0.89453125
                },
                {
                    "start": 1724.1,
                    "end": 1724.32,
                    "word": " do",
                    "probability": 0.9013671875
                },
                {
                    "start": 1724.32,
                    "end": 1724.56,
                    "word": ".",
                    "probability": 0.92919921875
                }
            ]
        },
        {
            "id": 279,
            "seek": 173381,
            "start": 1724.56,
            "end": 1727.62,
            "text": " But of course, a lot of times when you're dealing with arbitrary data,",
            "tokens": [
                51214,
                583,
                295,
                1164,
                11,
                257,
                688,
                295,
                1413,
                562,
                291,
                434,
                6260,
                365,
                23211,
                1412,
                11,
                51314
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17625899516421256,
            "compression_ratio": 1.7491961414790997,
            "no_speech_prob": 0.042083740234375,
            "words": [
                {
                    "start": 1724.56,
                    "end": 1724.64,
                    "word": " But",
                    "probability": 0.888671875
                },
                {
                    "start": 1724.64,
                    "end": 1724.82,
                    "word": " of",
                    "probability": 0.75146484375
                },
                {
                    "start": 1724.82,
                    "end": 1725.02,
                    "word": " course",
                    "probability": 0.8671875
                },
                {
                    "start": 1725.02,
                    "end": 1725.1,
                    "word": ",",
                    "probability": 0.8701171875
                },
                {
                    "start": 1725.3,
                    "end": 1725.48,
                    "word": " a lot",
                    "probability": 0.8857421875
                },
                {
                    "start": 1725.48,
                    "end": 1725.58,
                    "word": " of",
                    "probability": 0.90625
                },
                {
                    "start": 1725.58,
                    "end": 1725.84,
                    "word": " times",
                    "probability": 0.81884765625
                },
                {
                    "start": 1725.84,
                    "end": 1726.02,
                    "word": " when",
                    "probability": 0.77294921875
                },
                {
                    "start": 1726.02,
                    "end": 1726.16,
                    "word": " you're",
                    "probability": 0.924072265625
                },
                {
                    "start": 1726.16,
                    "end": 1726.36,
                    "word": " dealing",
                    "probability": 0.81689453125
                },
                {
                    "start": 1726.36,
                    "end": 1726.6,
                    "word": " with",
                    "probability": 0.806640625
                },
                {
                    "start": 1726.6,
                    "end": 1726.98,
                    "word": " arbitrary",
                    "probability": 0.74658203125
                },
                {
                    "start": 1726.98,
                    "end": 1727.36,
                    "word": " data",
                    "probability": 0.80615234375
                },
                {
                    "start": 1727.36,
                    "end": 1727.62,
                    "word": ",",
                    "probability": 0.916015625
                }
            ]
        },
        {
            "id": 280,
            "seek": 173381,
            "start": 1727.62,
            "end": 1733.08,
            "text": " machine learning, that sort of thing, you have no idea what bounds that are in the data.",
            "tokens": [
                51364,
                3479,
                2539,
                11,
                300,
                1333,
                295,
                551,
                11,
                291,
                362,
                572,
                1558,
                437,
                29905,
                300,
                366,
                294,
                264,
                1412,
                13,
                51564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17625899516421256,
            "compression_ratio": 1.7491961414790997,
            "no_speech_prob": 0.042083740234375,
            "words": [
                {
                    "start": 1727.62,
                    "end": 1727.88,
                    "word": " machine",
                    "probability": 0.6884765625
                },
                {
                    "start": 1727.88,
                    "end": 1728.18,
                    "word": " learning",
                    "probability": 0.8388671875
                },
                {
                    "start": 1728.18,
                    "end": 1728.28,
                    "word": ",",
                    "probability": 0.92724609375
                },
                {
                    "start": 1728.28,
                    "end": 1728.4,
                    "word": " that",
                    "probability": 0.8828125
                },
                {
                    "start": 1728.4,
                    "end": 1728.58,
                    "word": " sort",
                    "probability": 0.81982421875
                },
                {
                    "start": 1728.58,
                    "end": 1728.7,
                    "word": " of",
                    "probability": 0.90673828125
                },
                {
                    "start": 1728.7,
                    "end": 1728.92,
                    "word": " thing",
                    "probability": 0.8173828125
                },
                {
                    "start": 1728.92,
                    "end": 1729.22,
                    "word": ",",
                    "probability": 0.9619140625
                },
                {
                    "start": 1729.22,
                    "end": 1729.32,
                    "word": " you",
                    "probability": 0.892578125
                },
                {
                    "start": 1729.32,
                    "end": 1729.44,
                    "word": " have",
                    "probability": 0.87939453125
                },
                {
                    "start": 1729.44,
                    "end": 1729.64,
                    "word": " no",
                    "probability": 0.90673828125
                },
                {
                    "start": 1729.64,
                    "end": 1730.02,
                    "word": " idea",
                    "probability": 0.77490234375
                },
                {
                    "start": 1730.02,
                    "end": 1730.28,
                    "word": " what",
                    "probability": 0.88232421875
                },
                {
                    "start": 1730.28,
                    "end": 1730.94,
                    "word": " bounds",
                    "probability": 0.69970703125
                },
                {
                    "start": 1730.94,
                    "end": 1731.44,
                    "word": " that",
                    "probability": 0.83056640625
                },
                {
                    "start": 1731.44,
                    "end": 1732.18,
                    "word": " are",
                    "probability": 0.88720703125
                },
                {
                    "start": 1732.18,
                    "end": 1732.38,
                    "word": " in",
                    "probability": 0.90673828125
                },
                {
                    "start": 1732.38,
                    "end": 1732.5,
                    "word": " the",
                    "probability": 0.81640625
                },
                {
                    "start": 1732.5,
                    "end": 1732.74,
                    "word": " data",
                    "probability": 0.8203125
                },
                {
                    "start": 1732.74,
                    "end": 1733.08,
                    "word": ".",
                    "probability": 0.99072265625
                }
            ]
        },
        {
            "id": 281,
            "seek": 173381,
            "start": 1733.08,
            "end": 1735.9,
            "text": " In fact, that's maybe the reason you're doing the analysis in the first place",
            "tokens": [
                51614,
                682,
                1186,
                11,
                300,
                311,
                1310,
                264,
                1778,
                291,
                434,
                884,
                264,
                5215,
                294,
                264,
                700,
                1081,
                51714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17625899516421256,
            "compression_ratio": 1.7491961414790997,
            "no_speech_prob": 0.042083740234375,
            "words": [
                {
                    "start": 1733.08,
                    "end": 1733.14,
                    "word": " In",
                    "probability": 0.90966796875
                },
                {
                    "start": 1733.14,
                    "end": 1733.26,
                    "word": " fact",
                    "probability": 0.814453125
                },
                {
                    "start": 1733.26,
                    "end": 1733.28,
                    "word": ",",
                    "probability": 0.96533203125
                },
                {
                    "start": 1733.28,
                    "end": 1733.6,
                    "word": " that's",
                    "probability": 0.914306640625
                },
                {
                    "start": 1733.6,
                    "end": 1733.9,
                    "word": " maybe",
                    "probability": 0.87646484375
                },
                {
                    "start": 1733.9,
                    "end": 1734.12,
                    "word": " the",
                    "probability": 0.8154296875
                },
                {
                    "start": 1734.12,
                    "end": 1734.3,
                    "word": " reason",
                    "probability": 0.89794921875
                },
                {
                    "start": 1734.3,
                    "end": 1734.46,
                    "word": " you're",
                    "probability": 0.924560546875
                },
                {
                    "start": 1734.46,
                    "end": 1734.62,
                    "word": " doing",
                    "probability": 0.912109375
                },
                {
                    "start": 1734.62,
                    "end": 1734.78,
                    "word": " the",
                    "probability": 0.8046875
                },
                {
                    "start": 1734.78,
                    "end": 1735.1,
                    "word": " analysis",
                    "probability": 0.75390625
                },
                {
                    "start": 1735.1,
                    "end": 1735.34,
                    "word": " in",
                    "probability": 0.8876953125
                },
                {
                    "start": 1735.34,
                    "end": 1735.44,
                    "word": " the",
                    "probability": 0.81396484375
                },
                {
                    "start": 1735.44,
                    "end": 1735.62,
                    "word": " first",
                    "probability": 0.81494140625
                },
                {
                    "start": 1735.62,
                    "end": 1735.9,
                    "word": " place",
                    "probability": 0.8212890625
                }
            ]
        },
        {
            "id": 282,
            "seek": 173381,
            "start": 1735.9,
            "end": 1737.66,
            "text": " is to learn a bit more about the data.",
            "tokens": [
                51764,
                307,
                281,
                1466,
                257,
                857,
                544,
                466,
                264,
                1412,
                13,
                51814
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17625899516421256,
            "compression_ratio": 1.7491961414790997,
            "no_speech_prob": 0.042083740234375,
            "words": [
                {
                    "start": 1735.9,
                    "end": 1736.04,
                    "word": " is",
                    "probability": 0.52783203125
                },
                {
                    "start": 1736.04,
                    "end": 1736.18,
                    "word": " to",
                    "probability": 0.8994140625
                },
                {
                    "start": 1736.18,
                    "end": 1736.38,
                    "word": " learn",
                    "probability": 0.91162109375
                },
                {
                    "start": 1736.5,
                    "end": 1736.6,
                    "word": " a bit",
                    "probability": 0.89892578125
                },
                {
                    "start": 1736.6,
                    "end": 1736.78,
                    "word": " more",
                    "probability": 0.87744140625
                },
                {
                    "start": 1736.78,
                    "end": 1736.98,
                    "word": " about",
                    "probability": 0.8134765625
                },
                {
                    "start": 1736.98,
                    "end": 1737.12,
                    "word": " the",
                    "probability": 0.8115234375
                },
                {
                    "start": 1737.12,
                    "end": 1737.36,
                    "word": " data",
                    "probability": 0.82470703125
                },
                {
                    "start": 1737.36,
                    "end": 1737.66,
                    "word": ".",
                    "probability": 0.9921875
                }
            ]
        },
        {
            "id": 283,
            "seek": 176201,
            "start": 1738.48,
            "end": 1744.43,
            "text": " And so one of the other contributions of this work is automatic bounds determination.",
            "tokens": [
                50364,
                400,
                370,
                472,
                295,
                264,
                661,
                15725,
                295,
                341,
                589,
                307,
                12509,
                29905,
                18432,
                13,
                50664
            ],
            "temperature": 0.0,
            "avg_logprob": -0.16237745054212271,
            "compression_ratio": 1.6598360655737705,
            "no_speech_prob": 0.00388336181640625,
            "words": [
                {
                    "start": 1738.48,
                    "end": 1739.06,
                    "word": " And",
                    "probability": 0.73388671875
                },
                {
                    "start": 1739.06,
                    "end": 1739.38,
                    "word": " so",
                    "probability": 0.89306640625
                },
                {
                    "start": 1739.38,
                    "end": 1739.96,
                    "word": " one",
                    "probability": 0.64453125
                },
                {
                    "start": 1739.96,
                    "end": 1740.08,
                    "word": " of",
                    "probability": 0.90673828125
                },
                {
                    "start": 1740.08,
                    "end": 1740.18,
                    "word": " the",
                    "probability": 0.82080078125
                },
                {
                    "start": 1740.18,
                    "end": 1740.36,
                    "word": " other",
                    "probability": 0.78076171875
                },
                {
                    "start": 1740.36,
                    "end": 1740.86,
                    "word": " contributions",
                    "probability": 0.90283203125
                },
                {
                    "start": 1740.86,
                    "end": 1741.34,
                    "word": " of",
                    "probability": 0.9033203125
                },
                {
                    "start": 1741.34,
                    "end": 1741.92,
                    "word": " this",
                    "probability": 0.80224609375
                },
                {
                    "start": 1741.92,
                    "end": 1742.26,
                    "word": " work",
                    "probability": 0.81884765625
                },
                {
                    "start": 1742.26,
                    "end": 1742.64,
                    "word": " is",
                    "probability": 0.88818359375
                },
                {
                    "start": 1742.64,
                    "end": 1743.1,
                    "word": " automatic",
                    "probability": 0.7607421875
                },
                {
                    "start": 1743.1,
                    "end": 1743.54,
                    "word": " bounds",
                    "probability": 0.6044921875
                },
                {
                    "start": 1743.54,
                    "end": 1744.14,
                    "word": " determination",
                    "probability": 0.83251953125
                },
                {
                    "start": 1744.14,
                    "end": 1744.43,
                    "word": ".",
                    "probability": 0.9580078125
                }
            ]
        },
        {
            "id": 284,
            "seek": 176201,
            "start": 1745.84,
            "end": 1750.8,
            "text": " You want to be able to sort of get the best possible bounding on your data",
            "tokens": [
                50714,
                509,
                528,
                281,
                312,
                1075,
                281,
                1333,
                295,
                483,
                264,
                1151,
                1944,
                5472,
                278,
                322,
                428,
                1412,
                50964
            ],
            "temperature": 0.0,
            "avg_logprob": -0.16237745054212271,
            "compression_ratio": 1.6598360655737705,
            "no_speech_prob": 0.00388336181640625,
            "words": [
                {
                    "start": 1745.84,
                    "end": 1746.42,
                    "word": " You",
                    "probability": 0.88623046875
                },
                {
                    "start": 1746.42,
                    "end": 1747.0,
                    "word": " want",
                    "probability": 0.779296875
                },
                {
                    "start": 1747.0,
                    "end": 1747.22,
                    "word": " to",
                    "probability": 0.90673828125
                },
                {
                    "start": 1747.22,
                    "end": 1747.38,
                    "word": " be",
                    "probability": 0.90380859375
                },
                {
                    "start": 1747.38,
                    "end": 1747.58,
                    "word": " able",
                    "probability": 0.9033203125
                },
                {
                    "start": 1747.58,
                    "end": 1747.8,
                    "word": " to",
                    "probability": 0.904296875
                },
                {
                    "start": 1747.8,
                    "end": 1747.98,
                    "word": " sort",
                    "probability": 0.6669921875
                },
                {
                    "start": 1747.98,
                    "end": 1748.12,
                    "word": " of",
                    "probability": 0.9033203125
                },
                {
                    "start": 1748.12,
                    "end": 1748.5,
                    "word": " get",
                    "probability": 0.89013671875
                },
                {
                    "start": 1748.5,
                    "end": 1748.8,
                    "word": " the",
                    "probability": 0.8115234375
                },
                {
                    "start": 1748.8,
                    "end": 1749.36,
                    "word": " best",
                    "probability": 0.81591796875
                },
                {
                    "start": 1749.36,
                    "end": 1749.82,
                    "word": " possible",
                    "probability": 0.86083984375
                },
                {
                    "start": 1749.82,
                    "end": 1750.26,
                    "word": " bounding",
                    "probability": 0.88671875
                },
                {
                    "start": 1750.26,
                    "end": 1750.38,
                    "word": " on",
                    "probability": 0.90869140625
                },
                {
                    "start": 1750.38,
                    "end": 1750.5,
                    "word": " your",
                    "probability": 0.80224609375
                },
                {
                    "start": 1750.5,
                    "end": 1750.8,
                    "word": " data",
                    "probability": 0.818359375
                }
            ]
        },
        {
            "id": 285,
            "seek": 176201,
            "start": 1750.8,
            "end": 1753.58,
            "text": " without paying for it very much.",
            "tokens": [
                51014,
                1553,
                6229,
                337,
                309,
                588,
                709,
                13,
                51064
            ],
            "temperature": 0.0,
            "avg_logprob": -0.16237745054212271,
            "compression_ratio": 1.6598360655737705,
            "no_speech_prob": 0.00388336181640625,
            "words": [
                {
                    "start": 1750.8,
                    "end": 1751.76,
                    "word": " without",
                    "probability": 0.74365234375
                },
                {
                    "start": 1751.76,
                    "end": 1752.14,
                    "word": " paying",
                    "probability": 0.83447265625
                },
                {
                    "start": 1752.14,
                    "end": 1752.36,
                    "word": " for",
                    "probability": 0.890625
                },
                {
                    "start": 1752.36,
                    "end": 1752.48,
                    "word": " it",
                    "probability": 0.904296875
                },
                {
                    "start": 1752.48,
                    "end": 1752.68,
                    "word": " very",
                    "probability": 0.728515625
                },
                {
                    "start": 1752.68,
                    "end": 1753.02,
                    "word": " much",
                    "probability": 0.81201171875
                },
                {
                    "start": 1753.02,
                    "end": 1753.58,
                    "word": ".",
                    "probability": 0.98291015625
                }
            ]
        },
        {
            "id": 286,
            "seek": 176201,
            "start": 1753.58,
            "end": 1757.12,
            "text": " Once again, finding out global bounds that are differentially private",
            "tokens": [
                51114,
                3443,
                797,
                11,
                5006,
                484,
                4338,
                29905,
                300,
                366,
                819,
                2270,
                4551,
                51264
            ],
            "temperature": 0.0,
            "avg_logprob": -0.16237745054212271,
            "compression_ratio": 1.6598360655737705,
            "no_speech_prob": 0.00388336181640625,
            "words": [
                {
                    "start": 1753.58,
                    "end": 1753.78,
                    "word": " Once",
                    "probability": 0.845703125
                },
                {
                    "start": 1753.78,
                    "end": 1754.08,
                    "word": " again",
                    "probability": 0.91796875
                },
                {
                    "start": 1754.08,
                    "end": 1754.32,
                    "word": ",",
                    "probability": 0.92578125
                },
                {
                    "start": 1754.32,
                    "end": 1754.6,
                    "word": " finding",
                    "probability": 0.7626953125
                },
                {
                    "start": 1754.6,
                    "end": 1754.86,
                    "word": " out",
                    "probability": 0.81591796875
                },
                {
                    "start": 1754.86,
                    "end": 1755.18,
                    "word": " global",
                    "probability": 0.85107421875
                },
                {
                    "start": 1755.18,
                    "end": 1755.68,
                    "word": " bounds",
                    "probability": 0.73779296875
                },
                {
                    "start": 1755.68,
                    "end": 1756.26,
                    "word": " that",
                    "probability": 0.83837890625
                },
                {
                    "start": 1756.26,
                    "end": 1756.36,
                    "word": " are",
                    "probability": 0.8974609375
                },
                {
                    "start": 1756.36,
                    "end": 1756.76,
                    "word": " differentially",
                    "probability": 0.850830078125
                },
                {
                    "start": 1756.76,
                    "end": 1757.12,
                    "word": " private",
                    "probability": 0.78662109375
                }
            ]
        },
        {
            "id": 287,
            "seek": 176201,
            "start": 1757.12,
            "end": 1760.66,
            "text": " on arbitrary functions is undecidable.",
            "tokens": [
                51314,
                322,
                23211,
                6828,
                307,
                674,
                3045,
                38089,
                13,
                51464
            ],
            "temperature": 0.0,
            "avg_logprob": -0.16237745054212271,
            "compression_ratio": 1.6598360655737705,
            "no_speech_prob": 0.00388336181640625,
            "words": [
                {
                    "start": 1757.12,
                    "end": 1758.14,
                    "word": " on",
                    "probability": 0.853515625
                },
                {
                    "start": 1758.14,
                    "end": 1758.48,
                    "word": " arbitrary",
                    "probability": 0.74853515625
                },
                {
                    "start": 1758.48,
                    "end": 1758.96,
                    "word": " functions",
                    "probability": 0.85693359375
                },
                {
                    "start": 1758.96,
                    "end": 1759.52,
                    "word": " is",
                    "probability": 0.8916015625
                },
                {
                    "start": 1759.52,
                    "end": 1760.08,
                    "word": " undecidable",
                    "probability": 0.8595377604166666
                },
                {
                    "start": 1760.08,
                    "end": 1760.66,
                    "word": ".",
                    "probability": 0.85107421875
                }
            ]
        },
        {
            "id": 288,
            "seek": 176201,
            "start": 1761.22,
            "end": 1763.98,
            "text": " So, you know, you can just sort of give up out of the gate",
            "tokens": [
                51514,
                407,
                11,
                291,
                458,
                11,
                291,
                393,
                445,
                1333,
                295,
                976,
                493,
                484,
                295,
                264,
                8539,
                51614
            ],
            "temperature": 0.0,
            "avg_logprob": -0.16237745054212271,
            "compression_ratio": 1.6598360655737705,
            "no_speech_prob": 0.00388336181640625,
            "words": [
                {
                    "start": 1761.22,
                    "end": 1761.62,
                    "word": " So",
                    "probability": 0.85888671875
                },
                {
                    "start": 1761.62,
                    "end": 1762.2,
                    "word": ",",
                    "probability": 0.587890625
                },
                {
                    "start": 1762.2,
                    "end": 1762.36,
                    "word": " you",
                    "probability": 0.89208984375
                },
                {
                    "start": 1762.36,
                    "end": 1762.42,
                    "word": " know",
                    "probability": 0.52001953125
                },
                {
                    "start": 1762.42,
                    "end": 1762.46,
                    "word": ",",
                    "probability": 0.958984375
                },
                {
                    "start": 1762.46,
                    "end": 1762.56,
                    "word": " you",
                    "probability": 0.89404296875
                },
                {
                    "start": 1762.56,
                    "end": 1762.72,
                    "word": " can",
                    "probability": 0.88916015625
                },
                {
                    "start": 1762.72,
                    "end": 1762.94,
                    "word": " just",
                    "probability": 0.81982421875
                },
                {
                    "start": 1762.94,
                    "end": 1763.1,
                    "word": " sort",
                    "probability": 0.80419921875
                },
                {
                    "start": 1763.1,
                    "end": 1763.18,
                    "word": " of",
                    "probability": 0.90380859375
                },
                {
                    "start": 1763.18,
                    "end": 1763.34,
                    "word": " give",
                    "probability": 0.81787109375
                },
                {
                    "start": 1763.34,
                    "end": 1763.48,
                    "word": " up",
                    "probability": 0.90283203125
                },
                {
                    "start": 1763.48,
                    "end": 1763.6,
                    "word": " out",
                    "probability": 0.798828125
                },
                {
                    "start": 1763.6,
                    "end": 1763.68,
                    "word": " of",
                    "probability": 0.84521484375
                },
                {
                    "start": 1763.68,
                    "end": 1763.8,
                    "word": " the",
                    "probability": 0.81640625
                },
                {
                    "start": 1763.8,
                    "end": 1763.98,
                    "word": " gate",
                    "probability": 0.86474609375
                }
            ]
        },
        {
            "id": 289,
            "seek": 176201,
            "start": 1763.98,
            "end": 1766.38,
            "text": " and just sort of go to the next best thing,",
            "tokens": [
                51664,
                293,
                445,
                1333,
                295,
                352,
                281,
                264,
                958,
                1151,
                551,
                11,
                51714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.16237745054212271,
            "compression_ratio": 1.6598360655737705,
            "no_speech_prob": 0.00388336181640625,
            "words": [
                {
                    "start": 1763.98,
                    "end": 1764.14,
                    "word": " and",
                    "probability": 0.87744140625
                },
                {
                    "start": 1764.14,
                    "end": 1764.34,
                    "word": " just",
                    "probability": 0.88623046875
                },
                {
                    "start": 1764.34,
                    "end": 1764.9,
                    "word": " sort",
                    "probability": 0.779296875
                },
                {
                    "start": 1764.9,
                    "end": 1764.98,
                    "word": " of",
                    "probability": 0.9013671875
                },
                {
                    "start": 1764.98,
                    "end": 1765.1,
                    "word": " go",
                    "probability": 0.90087890625
                },
                {
                    "start": 1765.1,
                    "end": 1765.2,
                    "word": " to",
                    "probability": 0.90576171875
                },
                {
                    "start": 1765.2,
                    "end": 1765.3,
                    "word": " the",
                    "probability": 0.81396484375
                },
                {
                    "start": 1765.3,
                    "end": 1765.44,
                    "word": " next",
                    "probability": 0.85498046875
                },
                {
                    "start": 1765.44,
                    "end": 1765.66,
                    "word": " best",
                    "probability": 0.814453125
                },
                {
                    "start": 1765.66,
                    "end": 1766.0,
                    "word": " thing",
                    "probability": 0.81787109375
                },
                {
                    "start": 1766.0,
                    "end": 1766.38,
                    "word": ",",
                    "probability": 0.7958984375
                }
            ]
        },
        {
            "id": 290,
            "seek": 178793,
            "start": 1766.38,
            "end": 1768.2,
            "text": " which is a good approximation.",
            "tokens": [
                50364,
                597,
                307,
                257,
                665,
                28023,
                13,
                50414
            ],
            "temperature": 0.0,
            "avg_logprob": -0.15531250432133675,
            "compression_ratio": 1.611764705882353,
            "no_speech_prob": 0.00638580322265625,
            "words": [
                {
                    "start": 1766.38,
                    "end": 1766.58,
                    "word": " which",
                    "probability": 0.73681640625
                },
                {
                    "start": 1766.58,
                    "end": 1766.76,
                    "word": " is",
                    "probability": 0.89404296875
                },
                {
                    "start": 1766.9,
                    "end": 1767.06,
                    "word": " a good",
                    "probability": 0.88232421875
                },
                {
                    "start": 1767.06,
                    "end": 1767.72,
                    "word": " approximation",
                    "probability": 0.78857421875
                },
                {
                    "start": 1767.72,
                    "end": 1768.2,
                    "word": ".",
                    "probability": 0.8876953125
                }
            ]
        },
        {
            "id": 291,
            "seek": 178793,
            "start": 1768.24,
            "end": 1773.44,
            "text": " And so what we use when bounds are not provided is logarithmic histograms.",
            "tokens": [
                50464,
                400,
                370,
                437,
                321,
                764,
                562,
                29905,
                366,
                406,
                5649,
                307,
                41473,
                355,
                13195,
                49816,
                82,
                13,
                50664
            ],
            "temperature": 0.0,
            "avg_logprob": -0.15531250432133675,
            "compression_ratio": 1.611764705882353,
            "no_speech_prob": 0.00638580322265625,
            "words": [
                {
                    "start": 1768.24,
                    "end": 1768.32,
                    "word": " And",
                    "probability": 0.8447265625
                },
                {
                    "start": 1768.32,
                    "end": 1768.52,
                    "word": " so",
                    "probability": 0.8994140625
                },
                {
                    "start": 1768.52,
                    "end": 1768.68,
                    "word": " what",
                    "probability": 0.79638671875
                },
                {
                    "start": 1768.68,
                    "end": 1768.84,
                    "word": " we",
                    "probability": 0.90673828125
                },
                {
                    "start": 1768.84,
                    "end": 1769.16,
                    "word": " use",
                    "probability": 0.830078125
                },
                {
                    "start": 1769.16,
                    "end": 1769.36,
                    "word": " when",
                    "probability": 0.85498046875
                },
                {
                    "start": 1769.36,
                    "end": 1769.62,
                    "word": " bounds",
                    "probability": 0.744140625
                },
                {
                    "start": 1769.62,
                    "end": 1769.7,
                    "word": " are",
                    "probability": 0.8955078125
                },
                {
                    "start": 1769.7,
                    "end": 1769.9,
                    "word": " not",
                    "probability": 0.892578125
                },
                {
                    "start": 1769.9,
                    "end": 1770.32,
                    "word": " provided",
                    "probability": 0.85205078125
                },
                {
                    "start": 1770.32,
                    "end": 1771.22,
                    "word": " is",
                    "probability": 0.849609375
                },
                {
                    "start": 1771.22,
                    "end": 1772.22,
                    "word": " logarithmic",
                    "probability": 0.8414713541666666
                },
                {
                    "start": 1772.22,
                    "end": 1773.24,
                    "word": " histograms",
                    "probability": 0.8369140625
                },
                {
                    "start": 1773.24,
                    "end": 1773.44,
                    "word": ".",
                    "probability": 0.90673828125
                }
            ]
        },
        {
            "id": 292,
            "seek": 178793,
            "start": 1773.44,
            "end": 1776.94,
            "text": " Basically, you have partitions of increasing size.",
            "tokens": [
                50714,
                8537,
                11,
                291,
                362,
                644,
                2451,
                295,
                5662,
                2744,
                13,
                50864
            ],
            "temperature": 0.0,
            "avg_logprob": -0.15531250432133675,
            "compression_ratio": 1.611764705882353,
            "no_speech_prob": 0.00638580322265625,
            "words": [
                {
                    "start": 1773.44,
                    "end": 1773.9,
                    "word": " Basically",
                    "probability": 0.91455078125
                },
                {
                    "start": 1773.9,
                    "end": 1774.4,
                    "word": ",",
                    "probability": 0.52001953125
                },
                {
                    "start": 1774.4,
                    "end": 1774.5,
                    "word": " you",
                    "probability": 0.8935546875
                },
                {
                    "start": 1774.5,
                    "end": 1774.92,
                    "word": " have",
                    "probability": 0.87255859375
                },
                {
                    "start": 1774.92,
                    "end": 1775.38,
                    "word": " partitions",
                    "probability": 0.918212890625
                },
                {
                    "start": 1775.38,
                    "end": 1775.56,
                    "word": " of",
                    "probability": 0.89404296875
                },
                {
                    "start": 1775.56,
                    "end": 1775.88,
                    "word": " increasing",
                    "probability": 0.91650390625
                },
                {
                    "start": 1775.88,
                    "end": 1776.46,
                    "word": " size",
                    "probability": 0.79296875
                },
                {
                    "start": 1776.46,
                    "end": 1776.94,
                    "word": ".",
                    "probability": 0.319091796875
                }
            ]
        },
        {
            "id": 293,
            "seek": 178793,
            "start": 1777.26,
            "end": 1782.62,
            "text": " Doubling, for instance, allows you to get a reasonably good upper bound",
            "tokens": [
                50914,
                13200,
                18262,
                11,
                337,
                5197,
                11,
                4045,
                291,
                281,
                483,
                257,
                23551,
                665,
                6597,
                5472,
                51164
            ],
            "temperature": 0.0,
            "avg_logprob": -0.15531250432133675,
            "compression_ratio": 1.611764705882353,
            "no_speech_prob": 0.00638580322265625,
            "words": [
                {
                    "start": 1777.26,
                    "end": 1777.48,
                    "word": " Doubling",
                    "probability": 0.923583984375
                },
                {
                    "start": 1777.48,
                    "end": 1777.62,
                    "word": ",",
                    "probability": 0.75732421875
                },
                {
                    "start": 1777.62,
                    "end": 1777.68,
                    "word": " for",
                    "probability": 0.89501953125
                },
                {
                    "start": 1777.68,
                    "end": 1778.12,
                    "word": " instance",
                    "probability": 0.92041015625
                },
                {
                    "start": 1778.12,
                    "end": 1779.12,
                    "word": ",",
                    "probability": 0.96484375
                },
                {
                    "start": 1779.12,
                    "end": 1779.48,
                    "word": " allows",
                    "probability": 0.8115234375
                },
                {
                    "start": 1779.48,
                    "end": 1779.66,
                    "word": " you",
                    "probability": 0.890625
                },
                {
                    "start": 1779.66,
                    "end": 1779.88,
                    "word": " to",
                    "probability": 0.90673828125
                },
                {
                    "start": 1779.88,
                    "end": 1780.26,
                    "word": " get",
                    "probability": 0.90234375
                },
                {
                    "start": 1780.88,
                    "end": 1781.42,
                    "word": " a reasonably",
                    "probability": 0.89111328125
                },
                {
                    "start": 1781.42,
                    "end": 1782.04,
                    "word": " good",
                    "probability": 0.86669921875
                },
                {
                    "start": 1782.04,
                    "end": 1782.28,
                    "word": " upper",
                    "probability": 0.82958984375
                },
                {
                    "start": 1782.28,
                    "end": 1782.62,
                    "word": " bound",
                    "probability": 0.8388671875
                }
            ]
        },
        {
            "id": 294,
            "seek": 178793,
            "start": 1782.62,
            "end": 1784.92,
            "text": " that might be off by a factor of two,",
            "tokens": [
                51214,
                300,
                1062,
                312,
                766,
                538,
                257,
                5952,
                295,
                732,
                11,
                51264
            ],
            "temperature": 0.0,
            "avg_logprob": -0.15531250432133675,
            "compression_ratio": 1.611764705882353,
            "no_speech_prob": 0.00638580322265625,
            "words": [
                {
                    "start": 1782.62,
                    "end": 1782.84,
                    "word": " that",
                    "probability": 0.8662109375
                },
                {
                    "start": 1782.84,
                    "end": 1783.0,
                    "word": " might",
                    "probability": 0.8115234375
                },
                {
                    "start": 1783.0,
                    "end": 1783.16,
                    "word": " be",
                    "probability": 0.8955078125
                },
                {
                    "start": 1783.16,
                    "end": 1783.4,
                    "word": " off",
                    "probability": 0.81884765625
                },
                {
                    "start": 1783.4,
                    "end": 1783.58,
                    "word": " by",
                    "probability": 0.90673828125
                },
                {
                    "start": 1783.72,
                    "end": 1783.96,
                    "word": " a factor",
                    "probability": 0.7412109375
                },
                {
                    "start": 1783.96,
                    "end": 1784.18,
                    "word": " of",
                    "probability": 0.9091796875
                },
                {
                    "start": 1784.18,
                    "end": 1784.46,
                    "word": " two",
                    "probability": 0.74072265625
                },
                {
                    "start": 1784.46,
                    "end": 1784.92,
                    "word": ",",
                    "probability": 0.8544921875
                }
            ]
        },
        {
            "id": 295,
            "seek": 178793,
            "start": 1784.92,
            "end": 1787.84,
            "text": " which in terms of differential privacy actually doesn't cost you very much",
            "tokens": [
                51314,
                597,
                294,
                2115,
                295,
                15756,
                11427,
                767,
                1177,
                380,
                2063,
                291,
                588,
                709,
                51414
            ],
            "temperature": 0.0,
            "avg_logprob": -0.15531250432133675,
            "compression_ratio": 1.611764705882353,
            "no_speech_prob": 0.00638580322265625,
            "words": [
                {
                    "start": 1784.92,
                    "end": 1785.2,
                    "word": " which",
                    "probability": 0.9130859375
                },
                {
                    "start": 1785.2,
                    "end": 1785.36,
                    "word": " in",
                    "probability": 0.82470703125
                },
                {
                    "start": 1785.36,
                    "end": 1785.58,
                    "word": " terms",
                    "probability": 0.77880859375
                },
                {
                    "start": 1785.58,
                    "end": 1785.74,
                    "word": " of",
                    "probability": 0.90380859375
                },
                {
                    "start": 1785.74,
                    "end": 1786.04,
                    "word": " differential",
                    "probability": 0.7275390625
                },
                {
                    "start": 1786.04,
                    "end": 1786.4,
                    "word": " privacy",
                    "probability": 0.85791015625
                },
                {
                    "start": 1786.4,
                    "end": 1786.7,
                    "word": " actually",
                    "probability": 0.501953125
                },
                {
                    "start": 1786.7,
                    "end": 1786.96,
                    "word": " doesn't",
                    "probability": 0.927001953125
                },
                {
                    "start": 1786.96,
                    "end": 1787.26,
                    "word": " cost",
                    "probability": 0.8154296875
                },
                {
                    "start": 1787.26,
                    "end": 1787.4,
                    "word": " you",
                    "probability": 0.8876953125
                },
                {
                    "start": 1787.4,
                    "end": 1787.58,
                    "word": " very",
                    "probability": 0.73193359375
                },
                {
                    "start": 1787.58,
                    "end": 1787.84,
                    "word": " much",
                    "probability": 0.82568359375
                }
            ]
        },
        {
            "id": 296,
            "seek": 178793,
            "start": 1787.84,
            "end": 1792.3,
            "text": " in terms of utility, but still gives you a really solid understanding",
            "tokens": [
                51464,
                294,
                2115,
                295,
                14877,
                11,
                457,
                920,
                2709,
                291,
                257,
                534,
                5100,
                3701,
                51664
            ],
            "temperature": 0.0,
            "avg_logprob": -0.15531250432133675,
            "compression_ratio": 1.611764705882353,
            "no_speech_prob": 0.00638580322265625,
            "words": [
                {
                    "start": 1787.84,
                    "end": 1787.98,
                    "word": " in",
                    "probability": 0.90087890625
                },
                {
                    "start": 1787.98,
                    "end": 1788.18,
                    "word": " terms",
                    "probability": 0.775390625
                },
                {
                    "start": 1788.18,
                    "end": 1788.3,
                    "word": " of",
                    "probability": 0.904296875
                },
                {
                    "start": 1788.3,
                    "end": 1788.72,
                    "word": " utility",
                    "probability": 0.84814453125
                },
                {
                    "start": 1788.72,
                    "end": 1789.26,
                    "word": ",",
                    "probability": 0.77734375
                },
                {
                    "start": 1789.26,
                    "end": 1789.46,
                    "word": " but",
                    "probability": 0.896484375
                },
                {
                    "start": 1789.46,
                    "end": 1789.66,
                    "word": " still",
                    "probability": 0.9052734375
                },
                {
                    "start": 1789.66,
                    "end": 1789.86,
                    "word": " gives",
                    "probability": 0.8359375
                },
                {
                    "start": 1789.86,
                    "end": 1789.98,
                    "word": " you",
                    "probability": 0.890625
                },
                {
                    "start": 1790.26,
                    "end": 1790.6,
                    "word": " a really",
                    "probability": 0.7822265625
                },
                {
                    "start": 1790.6,
                    "end": 1791.04,
                    "word": " solid",
                    "probability": 0.8525390625
                },
                {
                    "start": 1791.04,
                    "end": 1792.3,
                    "word": " understanding",
                    "probability": 0.70947265625
                }
            ]
        },
        {
            "id": 297,
            "seek": 181793,
            "start": 1792.3,
            "end": 1796.76,
            "text": " of, you know, sort of the worst case upper bound.",
            "tokens": [
                50364,
                295,
                11,
                291,
                458,
                11,
                1333,
                295,
                264,
                5855,
                1389,
                6597,
                5472,
                13,
                50514
            ],
            "temperature": 0.0,
            "avg_logprob": -0.213709678060265,
            "compression_ratio": 1.7531914893617022,
            "no_speech_prob": 0.018829345703125,
            "words": [
                {
                    "start": 1792.3,
                    "end": 1793.38,
                    "word": " of",
                    "probability": 0.59033203125
                },
                {
                    "start": 1793.38,
                    "end": 1793.48,
                    "word": ",",
                    "probability": 0.1446533203125
                },
                {
                    "start": 1793.48,
                    "end": 1794.34,
                    "word": " you",
                    "probability": 0.73193359375
                },
                {
                    "start": 1794.34,
                    "end": 1794.44,
                    "word": " know",
                    "probability": 0.81201171875
                },
                {
                    "start": 1794.44,
                    "end": 1794.48,
                    "word": ",",
                    "probability": 0.958984375
                },
                {
                    "start": 1794.48,
                    "end": 1794.62,
                    "word": " sort",
                    "probability": 0.81201171875
                },
                {
                    "start": 1794.62,
                    "end": 1794.76,
                    "word": " of",
                    "probability": 0.90380859375
                },
                {
                    "start": 1794.76,
                    "end": 1795.34,
                    "word": " the",
                    "probability": 0.78125
                },
                {
                    "start": 1795.34,
                    "end": 1795.82,
                    "word": " worst",
                    "probability": 0.77734375
                },
                {
                    "start": 1795.82,
                    "end": 1796.1,
                    "word": " case",
                    "probability": 0.67333984375
                },
                {
                    "start": 1796.1,
                    "end": 1796.32,
                    "word": " upper",
                    "probability": 0.806640625
                },
                {
                    "start": 1796.32,
                    "end": 1796.64,
                    "word": " bound",
                    "probability": 0.87841796875
                },
                {
                    "start": 1796.64,
                    "end": 1796.76,
                    "word": ".",
                    "probability": 0.8955078125
                }
            ]
        },
        {
            "id": 298,
            "seek": 181793,
            "start": 1796.76,
            "end": 1801.38,
            "text": " And what's really amazing about this is that, you know,",
            "tokens": [
                50564,
                400,
                437,
                311,
                534,
                2243,
                466,
                341,
                307,
                300,
                11,
                291,
                458,
                11,
                50764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.213709678060265,
            "compression_ratio": 1.7531914893617022,
            "no_speech_prob": 0.018829345703125,
            "words": [
                {
                    "start": 1796.76,
                    "end": 1796.84,
                    "word": " And",
                    "probability": 0.89306640625
                },
                {
                    "start": 1796.84,
                    "end": 1797.02,
                    "word": " what's",
                    "probability": 0.9423828125
                },
                {
                    "start": 1797.02,
                    "end": 1797.18,
                    "word": " really",
                    "probability": 0.77978515625
                },
                {
                    "start": 1797.18,
                    "end": 1797.66,
                    "word": " amazing",
                    "probability": 0.8369140625
                },
                {
                    "start": 1797.66,
                    "end": 1797.94,
                    "word": " about",
                    "probability": 0.81787109375
                },
                {
                    "start": 1797.94,
                    "end": 1798.3,
                    "word": " this",
                    "probability": 0.88232421875
                },
                {
                    "start": 1798.3,
                    "end": 1799.56,
                    "word": " is",
                    "probability": 0.79931640625
                },
                {
                    "start": 1799.56,
                    "end": 1800.06,
                    "word": " that",
                    "probability": 0.880859375
                },
                {
                    "start": 1800.06,
                    "end": 1800.24,
                    "word": ",",
                    "probability": 0.86328125
                },
                {
                    "start": 1800.24,
                    "end": 1801.2,
                    "word": " you",
                    "probability": 0.8486328125
                },
                {
                    "start": 1801.2,
                    "end": 1801.3,
                    "word": " know",
                    "probability": 0.80810546875
                },
                {
                    "start": 1801.3,
                    "end": 1801.38,
                    "word": ",",
                    "probability": 0.94775390625
                }
            ]
        },
        {
            "id": 299,
            "seek": 181793,
            "start": 1801.38,
            "end": 1805.24,
            "text": " even with our epsilon having to be split between the exploration bounds",
            "tokens": [
                50814,
                754,
                365,
                527,
                17889,
                1419,
                281,
                312,
                7472,
                1296,
                264,
                16197,
                29905,
                50964
            ],
            "temperature": 0.0,
            "avg_logprob": -0.213709678060265,
            "compression_ratio": 1.7531914893617022,
            "no_speech_prob": 0.018829345703125,
            "words": [
                {
                    "start": 1801.38,
                    "end": 1801.66,
                    "word": " even",
                    "probability": 0.810546875
                },
                {
                    "start": 1801.66,
                    "end": 1802.02,
                    "word": " with",
                    "probability": 0.81005859375
                },
                {
                    "start": 1802.02,
                    "end": 1802.26,
                    "word": " our",
                    "probability": 0.70703125
                },
                {
                    "start": 1802.26,
                    "end": 1802.5,
                    "word": " epsilon",
                    "probability": 0.6416015625
                },
                {
                    "start": 1802.5,
                    "end": 1802.84,
                    "word": " having",
                    "probability": 0.76171875
                },
                {
                    "start": 1802.84,
                    "end": 1803.0,
                    "word": " to",
                    "probability": 0.90673828125
                },
                {
                    "start": 1803.0,
                    "end": 1803.16,
                    "word": " be",
                    "probability": 0.90283203125
                },
                {
                    "start": 1803.16,
                    "end": 1803.44,
                    "word": " split",
                    "probability": 0.84423828125
                },
                {
                    "start": 1803.44,
                    "end": 1803.88,
                    "word": " between",
                    "probability": 0.77685546875
                },
                {
                    "start": 1803.88,
                    "end": 1804.3,
                    "word": " the",
                    "probability": 0.80615234375
                },
                {
                    "start": 1804.3,
                    "end": 1804.74,
                    "word": " exploration",
                    "probability": 0.80712890625
                },
                {
                    "start": 1804.74,
                    "end": 1805.24,
                    "word": " bounds",
                    "probability": 0.75732421875
                }
            ]
        },
        {
            "id": 300,
            "seek": 181793,
            "start": 1805.24,
            "end": 1809.06,
            "text": " of the histogram and the actual computation,",
            "tokens": [
                51014,
                295,
                264,
                49816,
                293,
                264,
                3539,
                24903,
                11,
                51114
            ],
            "temperature": 0.0,
            "avg_logprob": -0.213709678060265,
            "compression_ratio": 1.7531914893617022,
            "no_speech_prob": 0.018829345703125,
            "words": [
                {
                    "start": 1805.24,
                    "end": 1805.78,
                    "word": " of",
                    "probability": 0.900390625
                },
                {
                    "start": 1805.78,
                    "end": 1806.16,
                    "word": " the",
                    "probability": 0.81005859375
                },
                {
                    "start": 1806.16,
                    "end": 1806.48,
                    "word": " histogram",
                    "probability": 0.7333984375
                },
                {
                    "start": 1806.48,
                    "end": 1807.22,
                    "word": " and",
                    "probability": 0.87548828125
                },
                {
                    "start": 1807.22,
                    "end": 1807.38,
                    "word": " the",
                    "probability": 0.82080078125
                },
                {
                    "start": 1807.38,
                    "end": 1807.62,
                    "word": " actual",
                    "probability": 0.7919921875
                },
                {
                    "start": 1807.62,
                    "end": 1808.24,
                    "word": " computation",
                    "probability": 0.8251953125
                },
                {
                    "start": 1808.24,
                    "end": 1809.06,
                    "word": ",",
                    "probability": 0.90283203125
                }
            ]
        },
        {
            "id": 301,
            "seek": 181793,
            "start": 1809.06,
            "end": 1812.9,
            "text": " the accuracy is often still better than manually determined bounds.",
            "tokens": [
                51164,
                264,
                14170,
                307,
                2049,
                920,
                1101,
                813,
                16945,
                9540,
                29905,
                13,
                51364
            ],
            "temperature": 0.0,
            "avg_logprob": -0.213709678060265,
            "compression_ratio": 1.7531914893617022,
            "no_speech_prob": 0.018829345703125,
            "words": [
                {
                    "start": 1809.06,
                    "end": 1809.18,
                    "word": " the",
                    "probability": 0.79736328125
                },
                {
                    "start": 1809.18,
                    "end": 1809.7,
                    "word": " accuracy",
                    "probability": 0.4375
                },
                {
                    "start": 1809.7,
                    "end": 1810.06,
                    "word": " is",
                    "probability": 0.90087890625
                },
                {
                    "start": 1810.06,
                    "end": 1810.34,
                    "word": " often",
                    "probability": 0.73681640625
                },
                {
                    "start": 1810.34,
                    "end": 1811.02,
                    "word": " still",
                    "probability": 0.90966796875
                },
                {
                    "start": 1811.02,
                    "end": 1811.28,
                    "word": " better",
                    "probability": 0.828125
                },
                {
                    "start": 1811.28,
                    "end": 1811.5,
                    "word": " than",
                    "probability": 0.88037109375
                },
                {
                    "start": 1811.5,
                    "end": 1811.84,
                    "word": " manually",
                    "probability": 0.83740234375
                },
                {
                    "start": 1811.84,
                    "end": 1812.26,
                    "word": " determined",
                    "probability": 0.79345703125
                },
                {
                    "start": 1812.26,
                    "end": 1812.7,
                    "word": " bounds",
                    "probability": 0.71533203125
                },
                {
                    "start": 1812.7,
                    "end": 1812.9,
                    "word": ".",
                    "probability": 0.880859375
                }
            ]
        },
        {
            "id": 302,
            "seek": 181793,
            "start": 1813.72,
            "end": 1818.46,
            "text": " Again, just because you're oftentimes talking about worst case scenarios",
            "tokens": [
                51414,
                3764,
                11,
                445,
                570,
                291,
                434,
                18349,
                1417,
                466,
                5855,
                1389,
                15077,
                51614
            ],
            "temperature": 0.0,
            "avg_logprob": -0.213709678060265,
            "compression_ratio": 1.7531914893617022,
            "no_speech_prob": 0.018829345703125,
            "words": [
                {
                    "start": 1813.72,
                    "end": 1814.36,
                    "word": " Again",
                    "probability": 0.900390625
                },
                {
                    "start": 1814.36,
                    "end": 1814.44,
                    "word": ",",
                    "probability": 0.95556640625
                },
                {
                    "start": 1814.44,
                    "end": 1814.6,
                    "word": " just",
                    "probability": 0.88720703125
                },
                {
                    "start": 1814.6,
                    "end": 1815.12,
                    "word": " because",
                    "probability": 0.83740234375
                },
                {
                    "start": 1815.12,
                    "end": 1816.52,
                    "word": " you're",
                    "probability": 0.896728515625
                },
                {
                    "start": 1816.52,
                    "end": 1816.88,
                    "word": " oftentimes",
                    "probability": 0.693359375
                },
                {
                    "start": 1816.88,
                    "end": 1817.4,
                    "word": " talking",
                    "probability": 0.7470703125
                },
                {
                    "start": 1817.4,
                    "end": 1817.6,
                    "word": " about",
                    "probability": 0.82275390625
                },
                {
                    "start": 1817.6,
                    "end": 1817.86,
                    "word": " worst",
                    "probability": 0.8984375
                },
                {
                    "start": 1817.86,
                    "end": 1818.08,
                    "word": " case",
                    "probability": 0.7529296875
                },
                {
                    "start": 1818.08,
                    "end": 1818.46,
                    "word": " scenarios",
                    "probability": 0.82275390625
                }
            ]
        },
        {
            "id": 303,
            "seek": 181793,
            "start": 1818.46,
            "end": 1821.34,
            "text": " when you're determining bounds in a computation.",
            "tokens": [
                51664,
                562,
                291,
                434,
                23751,
                29905,
                294,
                257,
                24903,
                13,
                51814
            ],
            "temperature": 0.0,
            "avg_logprob": -0.213709678060265,
            "compression_ratio": 1.7531914893617022,
            "no_speech_prob": 0.018829345703125,
            "words": [
                {
                    "start": 1818.46,
                    "end": 1818.78,
                    "word": " when",
                    "probability": 0.81640625
                },
                {
                    "start": 1818.78,
                    "end": 1818.92,
                    "word": " you're",
                    "probability": 0.9248046875
                },
                {
                    "start": 1818.92,
                    "end": 1819.26,
                    "word": " determining",
                    "probability": 0.822265625
                },
                {
                    "start": 1819.26,
                    "end": 1819.72,
                    "word": " bounds",
                    "probability": 0.75732421875
                },
                {
                    "start": 1819.72,
                    "end": 1820.04,
                    "word": " in",
                    "probability": 0.90185546875
                },
                {
                    "start": 1820.52,
                    "end": 1821.34,
                    "word": " a computation",
                    "probability": 0.74462890625
                },
                {
                    "start": 1821.34,
                    "end": 1821.34,
                    "word": ".",
                    "probability": 0.6826171875
                }
            ]
        },
        {
            "id": 304,
            "seek": 184699,
            "start": 1822.3,
            "end": 1826.18,
            "text": " And this is finding them sort of data dependent bounds.",
            "tokens": [
                50364,
                400,
                341,
                307,
                5006,
                552,
                1333,
                295,
                1412,
                12334,
                29905,
                13,
                50514
            ],
            "temperature": 0.0,
            "avg_logprob": -0.21280570765552315,
            "compression_ratio": 1.5375,
            "no_speech_prob": 0.0009546279907226562,
            "words": [
                {
                    "start": 1822.3,
                    "end": 1822.52,
                    "word": " And",
                    "probability": 0.57568359375
                },
                {
                    "start": 1822.52,
                    "end": 1822.7,
                    "word": " this",
                    "probability": 0.87158203125
                },
                {
                    "start": 1822.7,
                    "end": 1822.84,
                    "word": " is",
                    "probability": 0.90576171875
                },
                {
                    "start": 1822.84,
                    "end": 1823.16,
                    "word": " finding",
                    "probability": 0.7431640625
                },
                {
                    "start": 1823.16,
                    "end": 1823.54,
                    "word": " them",
                    "probability": 0.73974609375
                },
                {
                    "start": 1823.54,
                    "end": 1823.86,
                    "word": " sort",
                    "probability": 0.50537109375
                },
                {
                    "start": 1823.86,
                    "end": 1824.08,
                    "word": " of",
                    "probability": 0.9052734375
                },
                {
                    "start": 1824.08,
                    "end": 1824.82,
                    "word": " data",
                    "probability": 0.7568359375
                },
                {
                    "start": 1824.82,
                    "end": 1825.2,
                    "word": " dependent",
                    "probability": 0.461669921875
                },
                {
                    "start": 1825.2,
                    "end": 1825.54,
                    "word": " bounds",
                    "probability": 0.81201171875
                },
                {
                    "start": 1825.54,
                    "end": 1826.18,
                    "word": ".",
                    "probability": 0.93115234375
                }
            ]
        },
        {
            "id": 305,
            "seek": 184699,
            "start": 1826.52,
            "end": 1830.16,
            "text": " And typically, you know, you might require maybe 100 elements",
            "tokens": [
                50564,
                400,
                5850,
                11,
                291,
                458,
                11,
                291,
                1062,
                3651,
                1310,
                2319,
                4959,
                50714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.21280570765552315,
            "compression_ratio": 1.5375,
            "no_speech_prob": 0.0009546279907226562,
            "words": [
                {
                    "start": 1826.52,
                    "end": 1827.08,
                    "word": " And",
                    "probability": 0.86962890625
                },
                {
                    "start": 1827.08,
                    "end": 1827.44,
                    "word": " typically",
                    "probability": 0.8505859375
                },
                {
                    "start": 1827.44,
                    "end": 1827.7,
                    "word": ",",
                    "probability": 0.88427734375
                },
                {
                    "start": 1827.7,
                    "end": 1827.8,
                    "word": " you",
                    "probability": 0.8896484375
                },
                {
                    "start": 1827.8,
                    "end": 1827.88,
                    "word": " know",
                    "probability": 0.411865234375
                },
                {
                    "start": 1827.88,
                    "end": 1827.94,
                    "word": ",",
                    "probability": 0.96044921875
                },
                {
                    "start": 1827.94,
                    "end": 1828.22,
                    "word": " you",
                    "probability": 0.8935546875
                },
                {
                    "start": 1828.22,
                    "end": 1828.5,
                    "word": " might",
                    "probability": 0.80078125
                },
                {
                    "start": 1828.5,
                    "end": 1829.06,
                    "word": " require",
                    "probability": 0.8349609375
                },
                {
                    "start": 1829.06,
                    "end": 1829.3,
                    "word": " maybe",
                    "probability": 0.85009765625
                },
                {
                    "start": 1829.3,
                    "end": 1829.56,
                    "word": " 100",
                    "probability": 0.837890625
                },
                {
                    "start": 1829.56,
                    "end": 1830.16,
                    "word": " elements",
                    "probability": 0.810546875
                }
            ]
        },
        {
            "id": 306,
            "seek": 184699,
            "start": 1830.16,
            "end": 1834.58,
            "text": " or fewer, depending on the nature of the calculation.",
            "tokens": [
                50764,
                420,
                13366,
                11,
                5413,
                322,
                264,
                3687,
                295,
                264,
                17108,
                13,
                50964
            ],
            "temperature": 0.0,
            "avg_logprob": -0.21280570765552315,
            "compression_ratio": 1.5375,
            "no_speech_prob": 0.0009546279907226562,
            "words": [
                {
                    "start": 1830.16,
                    "end": 1831.0,
                    "word": " or",
                    "probability": 0.70458984375
                },
                {
                    "start": 1831.0,
                    "end": 1831.7,
                    "word": " fewer",
                    "probability": 0.7900390625
                },
                {
                    "start": 1831.7,
                    "end": 1832.08,
                    "word": ",",
                    "probability": 0.6728515625
                },
                {
                    "start": 1832.08,
                    "end": 1832.5,
                    "word": " depending",
                    "probability": 0.81787109375
                },
                {
                    "start": 1832.5,
                    "end": 1832.8,
                    "word": " on",
                    "probability": 0.9091796875
                },
                {
                    "start": 1832.8,
                    "end": 1833.04,
                    "word": " the",
                    "probability": 0.81640625
                },
                {
                    "start": 1833.04,
                    "end": 1833.26,
                    "word": " nature",
                    "probability": 0.79931640625
                },
                {
                    "start": 1833.26,
                    "end": 1833.38,
                    "word": " of",
                    "probability": 0.90869140625
                },
                {
                    "start": 1833.38,
                    "end": 1833.46,
                    "word": " the",
                    "probability": 0.81787109375
                },
                {
                    "start": 1833.46,
                    "end": 1833.94,
                    "word": " calculation",
                    "probability": 0.82177734375
                },
                {
                    "start": 1833.94,
                    "end": 1834.58,
                    "word": ".",
                    "probability": 0.99609375
                }
            ]
        },
        {
            "id": 307,
            "seek": 184699,
            "start": 1836.0,
            "end": 1839.86,
            "text": " Now, let me go back to our open sourcing availability",
            "tokens": [
                51014,
                823,
                11,
                718,
                385,
                352,
                646,
                281,
                527,
                1269,
                11006,
                2175,
                17945,
                51214
            ],
            "temperature": 0.0,
            "avg_logprob": -0.21280570765552315,
            "compression_ratio": 1.5375,
            "no_speech_prob": 0.0009546279907226562,
            "words": [
                {
                    "start": 1836.0,
                    "end": 1836.58,
                    "word": " Now",
                    "probability": 0.72509765625
                },
                {
                    "start": 1836.58,
                    "end": 1836.8,
                    "word": ",",
                    "probability": 0.91552734375
                },
                {
                    "start": 1836.8,
                    "end": 1837.4,
                    "word": " let",
                    "probability": 0.8935546875
                },
                {
                    "start": 1837.4,
                    "end": 1837.48,
                    "word": " me",
                    "probability": 0.9013671875
                },
                {
                    "start": 1837.48,
                    "end": 1837.64,
                    "word": " go",
                    "probability": 0.8984375
                },
                {
                    "start": 1837.64,
                    "end": 1837.92,
                    "word": " back",
                    "probability": 0.81982421875
                },
                {
                    "start": 1837.92,
                    "end": 1838.26,
                    "word": " to",
                    "probability": 0.90673828125
                },
                {
                    "start": 1838.26,
                    "end": 1838.62,
                    "word": " our",
                    "probability": 0.8056640625
                },
                {
                    "start": 1838.62,
                    "end": 1839.0,
                    "word": " open",
                    "probability": 0.78173828125
                },
                {
                    "start": 1839.0,
                    "end": 1839.32,
                    "word": " sourcing",
                    "probability": 0.898681640625
                },
                {
                    "start": 1839.32,
                    "end": 1839.86,
                    "word": " availability",
                    "probability": 0.92333984375
                }
            ]
        },
        {
            "id": 308,
            "seek": 184699,
            "start": 1839.86,
            "end": 1843.28,
            "text": " and some action items for you.",
            "tokens": [
                51264,
                293,
                512,
                3069,
                4754,
                337,
                291,
                13,
                51414
            ],
            "temperature": 0.0,
            "avg_logprob": -0.21280570765552315,
            "compression_ratio": 1.5375,
            "no_speech_prob": 0.0009546279907226562,
            "words": [
                {
                    "start": 1839.86,
                    "end": 1840.7,
                    "word": " and",
                    "probability": 0.8564453125
                },
                {
                    "start": 1840.7,
                    "end": 1841.3,
                    "word": " some",
                    "probability": 0.7763671875
                },
                {
                    "start": 1841.3,
                    "end": 1841.7,
                    "word": " action",
                    "probability": 0.75
                },
                {
                    "start": 1841.7,
                    "end": 1842.18,
                    "word": " items",
                    "probability": 0.798828125
                },
                {
                    "start": 1842.18,
                    "end": 1842.38,
                    "word": " for",
                    "probability": 0.8974609375
                },
                {
                    "start": 1842.38,
                    "end": 1842.64,
                    "word": " you",
                    "probability": 0.89404296875
                },
                {
                    "start": 1842.64,
                    "end": 1843.28,
                    "word": ".",
                    "probability": 0.990234375
                }
            ]
        },
        {
            "id": 309,
            "seek": 184699,
            "start": 1845.02,
            "end": 1849.12,
            "text": " First, we would like to take part of our differentially private training",
            "tokens": [
                51464,
                2386,
                11,
                321,
                576,
                411,
                281,
                747,
                644,
                295,
                527,
                819,
                2270,
                4551,
                3097,
                51664
            ],
            "temperature": 0.0,
            "avg_logprob": -0.21280570765552315,
            "compression_ratio": 1.5375,
            "no_speech_prob": 0.0009546279907226562,
            "words": [
                {
                    "start": 1845.02,
                    "end": 1845.66,
                    "word": " First",
                    "probability": 0.84130859375
                },
                {
                    "start": 1845.66,
                    "end": 1845.9,
                    "word": ",",
                    "probability": 0.9814453125
                },
                {
                    "start": 1845.9,
                    "end": 1846.06,
                    "word": " we",
                    "probability": 0.86376953125
                },
                {
                    "start": 1846.06,
                    "end": 1846.2,
                    "word": " would",
                    "probability": 0.8271484375
                },
                {
                    "start": 1846.2,
                    "end": 1846.4,
                    "word": " like",
                    "probability": 0.8798828125
                },
                {
                    "start": 1846.4,
                    "end": 1846.56,
                    "word": " to",
                    "probability": 0.5859375
                },
                {
                    "start": 1846.56,
                    "end": 1846.78,
                    "word": " take",
                    "probability": 0.80859375
                },
                {
                    "start": 1846.78,
                    "end": 1847.16,
                    "word": " part",
                    "probability": 0.74072265625
                },
                {
                    "start": 1847.16,
                    "end": 1847.54,
                    "word": " of",
                    "probability": 0.90283203125
                },
                {
                    "start": 1847.54,
                    "end": 1847.76,
                    "word": " our",
                    "probability": 0.80078125
                },
                {
                    "start": 1847.76,
                    "end": 1848.3,
                    "word": " differentially",
                    "probability": 0.802001953125
                },
                {
                    "start": 1848.3,
                    "end": 1848.62,
                    "word": " private",
                    "probability": 0.65771484375
                },
                {
                    "start": 1848.62,
                    "end": 1849.12,
                    "word": " training",
                    "probability": 0.79541015625
                }
            ]
        },
        {
            "id": 310,
            "seek": 184699,
            "start": 1849.12,
            "end": 1851.36,
            "text": " that we're piloting with this audience.",
            "tokens": [
                51714,
                300,
                321,
                434,
                9691,
                278,
                365,
                341,
                4034,
                13,
                51764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.21280570765552315,
            "compression_ratio": 1.5375,
            "no_speech_prob": 0.0009546279907226562,
            "words": [
                {
                    "start": 1849.12,
                    "end": 1849.36,
                    "word": " that",
                    "probability": 0.87744140625
                },
                {
                    "start": 1849.36,
                    "end": 1849.6,
                    "word": " we're",
                    "probability": 0.90234375
                },
                {
                    "start": 1849.6,
                    "end": 1850.08,
                    "word": " piloting",
                    "probability": 0.879638671875
                },
                {
                    "start": 1850.08,
                    "end": 1850.28,
                    "word": " with",
                    "probability": 0.818359375
                },
                {
                    "start": 1850.28,
                    "end": 1850.52,
                    "word": " this",
                    "probability": 0.55224609375
                },
                {
                    "start": 1850.52,
                    "end": 1850.98,
                    "word": " audience",
                    "probability": 0.8486328125
                },
                {
                    "start": 1850.98,
                    "end": 1851.36,
                    "word": ".",
                    "probability": 0.65576171875
                }
            ]
        },
        {
            "id": 311,
            "seek": 187483,
            "start": 1851.36,
            "end": 1853.84,
            "text": " As a first step, please take our survey.",
            "tokens": [
                50364,
                1018,
                257,
                700,
                1823,
                11,
                1767,
                747,
                527,
                8984,
                13,
                50464
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2521609144007906,
            "compression_ratio": 1.6434782608695653,
            "no_speech_prob": 0.0011692047119140625,
            "words": [
                {
                    "start": 1851.36,
                    "end": 1851.52,
                    "word": " As",
                    "probability": 0.77197265625
                },
                {
                    "start": 1851.64,
                    "end": 1851.88,
                    "word": " a first",
                    "probability": 0.8095703125
                },
                {
                    "start": 1851.88,
                    "end": 1852.26,
                    "word": " step",
                    "probability": 0.81005859375
                },
                {
                    "start": 1852.26,
                    "end": 1852.46,
                    "word": ",",
                    "probability": 0.95751953125
                },
                {
                    "start": 1852.46,
                    "end": 1852.7,
                    "word": " please",
                    "probability": 0.71630859375
                },
                {
                    "start": 1852.7,
                    "end": 1852.94,
                    "word": " take",
                    "probability": 0.7783203125
                },
                {
                    "start": 1852.94,
                    "end": 1853.2,
                    "word": " our",
                    "probability": 0.56103515625
                },
                {
                    "start": 1853.2,
                    "end": 1853.62,
                    "word": " survey",
                    "probability": 0.80810546875
                },
                {
                    "start": 1853.62,
                    "end": 1853.84,
                    "word": ".",
                    "probability": 0.1900634765625
                }
            ]
        },
        {
            "id": 312,
            "seek": 187483,
            "start": 1853.84,
            "end": 1855.76,
            "text": " And it only takes five minutes.",
            "tokens": [
                50514,
                400,
                309,
                787,
                2516,
                1732,
                2077,
                13,
                50564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2521609144007906,
            "compression_ratio": 1.6434782608695653,
            "no_speech_prob": 0.0011692047119140625,
            "words": [
                {
                    "start": 1853.84,
                    "end": 1853.96,
                    "word": " And",
                    "probability": 0.6865234375
                },
                {
                    "start": 1853.96,
                    "end": 1854.08,
                    "word": " it",
                    "probability": 0.892578125
                },
                {
                    "start": 1854.08,
                    "end": 1854.26,
                    "word": " only",
                    "probability": 0.8564453125
                },
                {
                    "start": 1854.26,
                    "end": 1854.62,
                    "word": " takes",
                    "probability": 0.72998046875
                },
                {
                    "start": 1854.62,
                    "end": 1854.94,
                    "word": " five",
                    "probability": 0.77197265625
                },
                {
                    "start": 1854.94,
                    "end": 1855.32,
                    "word": " minutes",
                    "probability": 0.87744140625
                },
                {
                    "start": 1855.32,
                    "end": 1855.76,
                    "word": ".",
                    "probability": 0.923828125
                }
            ]
        },
        {
            "id": 313,
            "seek": 187483,
            "start": 1855.76,
            "end": 1858.78,
            "text": " It tries to figure out how much have you understood",
            "tokens": [
                50614,
                467,
                9898,
                281,
                2573,
                484,
                577,
                709,
                362,
                291,
                7320,
                50714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2521609144007906,
            "compression_ratio": 1.6434782608695653,
            "no_speech_prob": 0.0011692047119140625,
            "words": [
                {
                    "start": 1855.76,
                    "end": 1856.04,
                    "word": " It",
                    "probability": 0.85791015625
                },
                {
                    "start": 1856.04,
                    "end": 1856.34,
                    "word": " tries",
                    "probability": 0.7451171875
                },
                {
                    "start": 1856.34,
                    "end": 1856.52,
                    "word": " to",
                    "probability": 0.90380859375
                },
                {
                    "start": 1856.52,
                    "end": 1856.68,
                    "word": " figure",
                    "probability": 0.91552734375
                },
                {
                    "start": 1856.68,
                    "end": 1856.96,
                    "word": " out",
                    "probability": 0.8291015625
                },
                {
                    "start": 1856.96,
                    "end": 1857.44,
                    "word": " how",
                    "probability": 0.83642578125
                },
                {
                    "start": 1857.44,
                    "end": 1857.68,
                    "word": " much",
                    "probability": 0.8193359375
                },
                {
                    "start": 1857.68,
                    "end": 1858.1,
                    "word": " have",
                    "probability": 0.77734375
                },
                {
                    "start": 1858.1,
                    "end": 1858.22,
                    "word": " you",
                    "probability": 0.87353515625
                },
                {
                    "start": 1858.22,
                    "end": 1858.78,
                    "word": " understood",
                    "probability": 0.87451171875
                }
            ]
        },
        {
            "id": 314,
            "seek": 187483,
            "start": 1858.78,
            "end": 1862.62,
            "text": " about differential privacy before you jump into the next topics.",
            "tokens": [
                50764,
                466,
                15756,
                11427,
                949,
                291,
                3012,
                666,
                264,
                958,
                8378,
                13,
                50914
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2521609144007906,
            "compression_ratio": 1.6434782608695653,
            "no_speech_prob": 0.0011692047119140625,
            "words": [
                {
                    "start": 1858.78,
                    "end": 1859.12,
                    "word": " about",
                    "probability": 0.56689453125
                },
                {
                    "start": 1859.12,
                    "end": 1859.48,
                    "word": " differential",
                    "probability": 0.697265625
                },
                {
                    "start": 1859.48,
                    "end": 1859.98,
                    "word": " privacy",
                    "probability": 0.66064453125
                },
                {
                    "start": 1859.98,
                    "end": 1861.1,
                    "word": " before",
                    "probability": 0.646484375
                },
                {
                    "start": 1861.1,
                    "end": 1861.32,
                    "word": " you",
                    "probability": 0.890625
                },
                {
                    "start": 1861.32,
                    "end": 1861.56,
                    "word": " jump",
                    "probability": 0.79736328125
                },
                {
                    "start": 1861.56,
                    "end": 1861.72,
                    "word": " into",
                    "probability": 0.72998046875
                },
                {
                    "start": 1861.72,
                    "end": 1861.88,
                    "word": " the",
                    "probability": 0.8134765625
                },
                {
                    "start": 1861.88,
                    "end": 1862.08,
                    "word": " next",
                    "probability": 0.8583984375
                },
                {
                    "start": 1862.08,
                    "end": 1862.34,
                    "word": " topics",
                    "probability": 0.61767578125
                },
                {
                    "start": 1862.34,
                    "end": 1862.62,
                    "word": ".",
                    "probability": 0.92333984375
                }
            ]
        },
        {
            "id": 315,
            "seek": 187483,
            "start": 1863.12,
            "end": 1865.46,
            "text": " Then you can check another introductory video",
            "tokens": [
                50964,
                1396,
                291,
                393,
                1520,
                1071,
                39048,
                960,
                51064
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2521609144007906,
            "compression_ratio": 1.6434782608695653,
            "no_speech_prob": 0.0011692047119140625,
            "words": [
                {
                    "start": 1863.12,
                    "end": 1863.44,
                    "word": " Then",
                    "probability": 0.81298828125
                },
                {
                    "start": 1863.44,
                    "end": 1863.68,
                    "word": " you",
                    "probability": 0.7265625
                },
                {
                    "start": 1863.68,
                    "end": 1863.8,
                    "word": " can",
                    "probability": 0.8896484375
                },
                {
                    "start": 1863.8,
                    "end": 1864.06,
                    "word": " check",
                    "probability": 0.8193359375
                },
                {
                    "start": 1864.06,
                    "end": 1864.48,
                    "word": " another",
                    "probability": 0.7470703125
                },
                {
                    "start": 1864.48,
                    "end": 1864.92,
                    "word": " introductory",
                    "probability": 0.72802734375
                },
                {
                    "start": 1864.92,
                    "end": 1865.46,
                    "word": " video",
                    "probability": 0.78173828125
                }
            ]
        },
        {
            "id": 316,
            "seek": 187483,
            "start": 1865.46,
            "end": 1868.76,
            "text": " on differential privacy in that link.",
            "tokens": [
                51114,
                322,
                15756,
                11427,
                294,
                300,
                2113,
                13,
                51214
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2521609144007906,
            "compression_ratio": 1.6434782608695653,
            "no_speech_prob": 0.0011692047119140625,
            "words": [
                {
                    "start": 1865.46,
                    "end": 1865.92,
                    "word": " on",
                    "probability": 0.896484375
                },
                {
                    "start": 1865.92,
                    "end": 1866.26,
                    "word": " differential",
                    "probability": 0.74560546875
                },
                {
                    "start": 1866.26,
                    "end": 1866.76,
                    "word": " privacy",
                    "probability": 0.63720703125
                },
                {
                    "start": 1866.76,
                    "end": 1867.6,
                    "word": " in",
                    "probability": 0.876953125
                },
                {
                    "start": 1867.6,
                    "end": 1867.84,
                    "word": " that",
                    "probability": 0.890625
                },
                {
                    "start": 1867.84,
                    "end": 1868.2,
                    "word": " link",
                    "probability": 0.89208984375
                },
                {
                    "start": 1868.2,
                    "end": 1868.76,
                    "word": ".",
                    "probability": 0.9970703125
                }
            ]
        },
        {
            "id": 317,
            "seek": 187483,
            "start": 1870.7,
            "end": 1873.86,
            "text": " Third, you can try out some of our code labs.",
            "tokens": [
                51264,
                12548,
                11,
                291,
                393,
                853,
                484,
                512,
                295,
                527,
                3089,
                20339,
                13,
                51464
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2521609144007906,
            "compression_ratio": 1.6434782608695653,
            "no_speech_prob": 0.0011692047119140625,
            "words": [
                {
                    "start": 1870.7,
                    "end": 1871.26,
                    "word": " Third",
                    "probability": 0.8525390625
                },
                {
                    "start": 1871.26,
                    "end": 1871.38,
                    "word": ",",
                    "probability": 0.9814453125
                },
                {
                    "start": 1871.38,
                    "end": 1871.52,
                    "word": " you",
                    "probability": 0.888671875
                },
                {
                    "start": 1871.52,
                    "end": 1871.68,
                    "word": " can",
                    "probability": 0.89697265625
                },
                {
                    "start": 1871.68,
                    "end": 1871.96,
                    "word": " try",
                    "probability": 0.8193359375
                },
                {
                    "start": 1871.96,
                    "end": 1872.22,
                    "word": " out",
                    "probability": 0.71728515625
                },
                {
                    "start": 1872.22,
                    "end": 1872.48,
                    "word": " some",
                    "probability": 0.8095703125
                },
                {
                    "start": 1872.48,
                    "end": 1872.62,
                    "word": " of",
                    "probability": 0.90283203125
                },
                {
                    "start": 1872.62,
                    "end": 1872.78,
                    "word": " our",
                    "probability": 0.8271484375
                },
                {
                    "start": 1872.78,
                    "end": 1873.02,
                    "word": " code",
                    "probability": 0.478271484375
                },
                {
                    "start": 1873.02,
                    "end": 1873.36,
                    "word": " labs",
                    "probability": 0.8681640625
                },
                {
                    "start": 1873.36,
                    "end": 1873.86,
                    "word": ".",
                    "probability": 0.96826171875
                }
            ]
        },
        {
            "id": 318,
            "seek": 187483,
            "start": 1873.86,
            "end": 1879.2,
            "text": " The link that's pointing out there points to a Go pipeline",
            "tokens": [
                51514,
                440,
                2113,
                300,
                311,
                12166,
                484,
                456,
                2793,
                281,
                257,
                1037,
                15517,
                51714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2521609144007906,
            "compression_ratio": 1.6434782608695653,
            "no_speech_prob": 0.0011692047119140625,
            "words": [
                {
                    "start": 1873.86,
                    "end": 1874.4,
                    "word": " The",
                    "probability": 0.8193359375
                },
                {
                    "start": 1874.4,
                    "end": 1874.58,
                    "word": " link",
                    "probability": 0.853515625
                },
                {
                    "start": 1874.58,
                    "end": 1875.34,
                    "word": " that's",
                    "probability": 0.5406494140625
                },
                {
                    "start": 1875.34,
                    "end": 1875.7,
                    "word": " pointing",
                    "probability": 0.75634765625
                },
                {
                    "start": 1875.7,
                    "end": 1875.94,
                    "word": " out",
                    "probability": 0.79150390625
                },
                {
                    "start": 1875.94,
                    "end": 1876.28,
                    "word": " there",
                    "probability": 0.85107421875
                },
                {
                    "start": 1876.28,
                    "end": 1877.2,
                    "word": " points",
                    "probability": 0.484375
                },
                {
                    "start": 1877.2,
                    "end": 1877.48,
                    "word": " to",
                    "probability": 0.8994140625
                },
                {
                    "start": 1877.96,
                    "end": 1878.66,
                    "word": " a Go",
                    "probability": 0.24169921875
                },
                {
                    "start": 1878.66,
                    "end": 1879.2,
                    "word": " pipeline",
                    "probability": 0.73388671875
                }
            ]
        },
        {
            "id": 319,
            "seek": 190313,
            "start": 1879.2,
            "end": 1883.14,
            "text": " that you can build to produce differentially private metrics",
            "tokens": [
                50364,
                300,
                291,
                393,
                1322,
                281,
                5258,
                819,
                2270,
                4551,
                16367,
                50514
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19419642758893443,
            "compression_ratio": 1.6609442060085837,
            "no_speech_prob": 0.0020503997802734375,
            "words": [
                {
                    "start": 1879.2,
                    "end": 1879.48,
                    "word": " that",
                    "probability": 0.76708984375
                },
                {
                    "start": 1879.48,
                    "end": 1879.62,
                    "word": " you",
                    "probability": 0.8974609375
                },
                {
                    "start": 1879.62,
                    "end": 1879.84,
                    "word": " can",
                    "probability": 0.888671875
                },
                {
                    "start": 1879.84,
                    "end": 1880.2,
                    "word": " build",
                    "probability": 0.25439453125
                },
                {
                    "start": 1880.2,
                    "end": 1880.78,
                    "word": " to",
                    "probability": 0.81591796875
                },
                {
                    "start": 1880.78,
                    "end": 1881.92,
                    "word": " produce",
                    "probability": 0.61474609375
                },
                {
                    "start": 1881.92,
                    "end": 1882.36,
                    "word": " differentially",
                    "probability": 0.5709228515625
                },
                {
                    "start": 1882.36,
                    "end": 1882.7,
                    "word": " private",
                    "probability": 0.63671875
                },
                {
                    "start": 1882.7,
                    "end": 1883.14,
                    "word": " metrics",
                    "probability": 0.7431640625
                }
            ]
        },
        {
            "id": 320,
            "seek": 190313,
            "start": 1883.14,
            "end": 1887.14,
            "text": " for a restaurant and its busy times.",
            "tokens": [
                50564,
                337,
                257,
                6383,
                293,
                1080,
                5856,
                1413,
                13,
                50714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19419642758893443,
            "compression_ratio": 1.6609442060085837,
            "no_speech_prob": 0.0020503997802734375,
            "words": [
                {
                    "start": 1883.14,
                    "end": 1883.78,
                    "word": " for",
                    "probability": 0.88232421875
                },
                {
                    "start": 1884.1,
                    "end": 1884.54,
                    "word": " a restaurant",
                    "probability": 0.79736328125
                },
                {
                    "start": 1884.54,
                    "end": 1885.16,
                    "word": " and",
                    "probability": 0.65966796875
                },
                {
                    "start": 1885.16,
                    "end": 1885.54,
                    "word": " its",
                    "probability": 0.447265625
                },
                {
                    "start": 1885.54,
                    "end": 1885.92,
                    "word": " busy",
                    "probability": 0.83740234375
                },
                {
                    "start": 1885.92,
                    "end": 1886.46,
                    "word": " times",
                    "probability": 0.8076171875
                },
                {
                    "start": 1886.46,
                    "end": 1887.14,
                    "word": ".",
                    "probability": 0.9658203125
                }
            ]
        },
        {
            "id": 321,
            "seek": 190313,
            "start": 1887.64,
            "end": 1890.38,
            "text": " Fourth, if you're curious about differential privacy",
            "tokens": [
                50764,
                23773,
                11,
                498,
                291,
                434,
                6369,
                466,
                15756,
                11427,
                50864
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19419642758893443,
            "compression_ratio": 1.6609442060085837,
            "no_speech_prob": 0.0020503997802734375,
            "words": [
                {
                    "start": 1887.64,
                    "end": 1888.22,
                    "word": " Fourth",
                    "probability": 0.82177734375
                },
                {
                    "start": 1888.22,
                    "end": 1888.34,
                    "word": ",",
                    "probability": 0.884765625
                },
                {
                    "start": 1888.34,
                    "end": 1888.42,
                    "word": " if",
                    "probability": 0.892578125
                },
                {
                    "start": 1888.42,
                    "end": 1888.62,
                    "word": " you're",
                    "probability": 0.9052734375
                },
                {
                    "start": 1888.62,
                    "end": 1889.14,
                    "word": " curious",
                    "probability": 0.79541015625
                },
                {
                    "start": 1889.14,
                    "end": 1889.48,
                    "word": " about",
                    "probability": 0.82568359375
                },
                {
                    "start": 1889.48,
                    "end": 1890.02,
                    "word": " differential",
                    "probability": 0.60693359375
                },
                {
                    "start": 1890.02,
                    "end": 1890.38,
                    "word": " privacy",
                    "probability": 0.591796875
                }
            ]
        },
        {
            "id": 322,
            "seek": 190313,
            "start": 1890.38,
            "end": 1894.02,
            "text": " and learning more, you can check out the blog that we're linking.",
            "tokens": [
                50914,
                293,
                2539,
                544,
                11,
                291,
                393,
                1520,
                484,
                264,
                6968,
                300,
                321,
                434,
                25775,
                13,
                51064
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19419642758893443,
            "compression_ratio": 1.6609442060085837,
            "no_speech_prob": 0.0020503997802734375,
            "words": [
                {
                    "start": 1890.38,
                    "end": 1890.62,
                    "word": " and",
                    "probability": 0.8583984375
                },
                {
                    "start": 1890.62,
                    "end": 1890.9,
                    "word": " learning",
                    "probability": 0.80126953125
                },
                {
                    "start": 1890.9,
                    "end": 1891.24,
                    "word": " more",
                    "probability": 0.88134765625
                },
                {
                    "start": 1891.24,
                    "end": 1891.34,
                    "word": ",",
                    "probability": 0.7724609375
                },
                {
                    "start": 1891.34,
                    "end": 1891.46,
                    "word": " you",
                    "probability": 0.89306640625
                },
                {
                    "start": 1891.46,
                    "end": 1891.62,
                    "word": " can",
                    "probability": 0.89599609375
                },
                {
                    "start": 1891.62,
                    "end": 1891.86,
                    "word": " check",
                    "probability": 0.84326171875
                },
                {
                    "start": 1891.86,
                    "end": 1892.06,
                    "word": " out",
                    "probability": 0.8125
                },
                {
                    "start": 1892.06,
                    "end": 1892.24,
                    "word": " the",
                    "probability": 0.80859375
                },
                {
                    "start": 1892.24,
                    "end": 1892.5,
                    "word": " blog",
                    "probability": 0.828125
                },
                {
                    "start": 1892.5,
                    "end": 1892.7,
                    "word": " that",
                    "probability": 0.87548828125
                },
                {
                    "start": 1892.7,
                    "end": 1893.0,
                    "word": " we're",
                    "probability": 0.921142578125
                },
                {
                    "start": 1893.0,
                    "end": 1893.34,
                    "word": " linking",
                    "probability": 0.74560546875
                },
                {
                    "start": 1893.34,
                    "end": 1894.02,
                    "word": ".",
                    "probability": 0.98291015625
                }
            ]
        },
        {
            "id": 323,
            "seek": 190313,
            "start": 1895.34,
            "end": 1900.9,
            "text": " Finally, provide us some feedback on the overall training course",
            "tokens": [
                51114,
                6288,
                11,
                2893,
                505,
                512,
                5824,
                322,
                264,
                4787,
                3097,
                1164,
                51414
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19419642758893443,
            "compression_ratio": 1.6609442060085837,
            "no_speech_prob": 0.0020503997802734375,
            "words": [
                {
                    "start": 1895.34,
                    "end": 1896.02,
                    "word": " Finally",
                    "probability": 0.701171875
                },
                {
                    "start": 1896.02,
                    "end": 1896.24,
                    "word": ",",
                    "probability": 0.94580078125
                },
                {
                    "start": 1896.24,
                    "end": 1896.56,
                    "word": " provide",
                    "probability": 0.89208984375
                },
                {
                    "start": 1896.56,
                    "end": 1896.78,
                    "word": " us",
                    "probability": 0.90380859375
                },
                {
                    "start": 1896.78,
                    "end": 1897.02,
                    "word": " some",
                    "probability": 0.81494140625
                },
                {
                    "start": 1897.02,
                    "end": 1897.58,
                    "word": " feedback",
                    "probability": 0.8486328125
                },
                {
                    "start": 1897.58,
                    "end": 1898.34,
                    "word": " on",
                    "probability": 0.908203125
                },
                {
                    "start": 1898.34,
                    "end": 1899.4,
                    "word": " the",
                    "probability": 0.8076171875
                },
                {
                    "start": 1899.4,
                    "end": 1899.9,
                    "word": " overall",
                    "probability": 0.85693359375
                },
                {
                    "start": 1899.9,
                    "end": 1900.4,
                    "word": " training",
                    "probability": 0.822265625
                },
                {
                    "start": 1900.4,
                    "end": 1900.9,
                    "word": " course",
                    "probability": 0.818359375
                }
            ]
        },
        {
            "id": 324,
            "seek": 190313,
            "start": 1900.9,
            "end": 1903.16,
            "text": " and the materials that we've created for you.",
            "tokens": [
                51464,
                293,
                264,
                5319,
                300,
                321,
                600,
                2942,
                337,
                291,
                13,
                51514
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19419642758893443,
            "compression_ratio": 1.6609442060085837,
            "no_speech_prob": 0.0020503997802734375,
            "words": [
                {
                    "start": 1900.9,
                    "end": 1901.08,
                    "word": " and",
                    "probability": 0.8798828125
                },
                {
                    "start": 1901.08,
                    "end": 1901.22,
                    "word": " the",
                    "probability": 0.8134765625
                },
                {
                    "start": 1901.22,
                    "end": 1901.6,
                    "word": " materials",
                    "probability": 0.666015625
                },
                {
                    "start": 1901.6,
                    "end": 1901.8,
                    "word": " that",
                    "probability": 0.873046875
                },
                {
                    "start": 1901.8,
                    "end": 1902.02,
                    "word": " we've",
                    "probability": 0.92431640625
                },
                {
                    "start": 1902.02,
                    "end": 1902.32,
                    "word": " created",
                    "probability": 0.744140625
                },
                {
                    "start": 1902.32,
                    "end": 1902.58,
                    "word": " for",
                    "probability": 0.8994140625
                },
                {
                    "start": 1902.58,
                    "end": 1902.76,
                    "word": " you",
                    "probability": 0.8935546875
                },
                {
                    "start": 1902.76,
                    "end": 1903.16,
                    "word": ".",
                    "probability": 0.97265625
                }
            ]
        },
        {
            "id": 325,
            "seek": 190313,
            "start": 1903.16,
            "end": 1907.5,
            "text": " We're really looking forward to hearing more from your end.",
            "tokens": [
                51564,
                492,
                434,
                534,
                1237,
                2128,
                281,
                4763,
                544,
                490,
                428,
                917,
                13,
                51714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.19419642758893443,
            "compression_ratio": 1.6609442060085837,
            "no_speech_prob": 0.0020503997802734375,
            "words": [
                {
                    "start": 1903.16,
                    "end": 1903.74,
                    "word": " We're",
                    "probability": 0.92626953125
                },
                {
                    "start": 1903.74,
                    "end": 1903.98,
                    "word": " really",
                    "probability": 0.77587890625
                },
                {
                    "start": 1903.98,
                    "end": 1904.3,
                    "word": " looking",
                    "probability": 0.80322265625
                },
                {
                    "start": 1904.3,
                    "end": 1904.66,
                    "word": " forward",
                    "probability": 0.84619140625
                },
                {
                    "start": 1904.66,
                    "end": 1905.22,
                    "word": " to",
                    "probability": 0.87841796875
                },
                {
                    "start": 1905.22,
                    "end": 1905.5,
                    "word": " hearing",
                    "probability": 0.79833984375
                },
                {
                    "start": 1905.5,
                    "end": 1905.92,
                    "word": " more",
                    "probability": 0.88232421875
                },
                {
                    "start": 1905.92,
                    "end": 1906.34,
                    "word": " from",
                    "probability": 0.80615234375
                },
                {
                    "start": 1906.34,
                    "end": 1906.56,
                    "word": " your",
                    "probability": 0.76611328125
                },
                {
                    "start": 1906.56,
                    "end": 1906.82,
                    "word": " end",
                    "probability": 0.78076171875
                },
                {
                    "start": 1906.82,
                    "end": 1907.5,
                    "word": ".",
                    "probability": 0.89599609375
                }
            ]
        },
        {
            "id": 326,
            "seek": 193313,
            "start": 1907.5,
            "end": 1911.92,
            "text": " And with that, we'll open it up for questions.",
            "tokens": [
                50364,
                400,
                365,
                300,
                11,
                321,
                603,
                1269,
                309,
                493,
                337,
                1651,
                13,
                50564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17412725601110374,
            "compression_ratio": 1.4754716981132074,
            "no_speech_prob": 0.0181732177734375,
            "words": [
                {
                    "start": 1907.5,
                    "end": 1908.24,
                    "word": " And",
                    "probability": 0.744140625
                },
                {
                    "start": 1908.24,
                    "end": 1908.54,
                    "word": " with",
                    "probability": 0.81005859375
                },
                {
                    "start": 1908.54,
                    "end": 1908.9,
                    "word": " that",
                    "probability": 0.88232421875
                },
                {
                    "start": 1908.9,
                    "end": 1909.2,
                    "word": ",",
                    "probability": 0.86669921875
                },
                {
                    "start": 1909.2,
                    "end": 1909.7,
                    "word": " we'll",
                    "probability": 0.917724609375
                },
                {
                    "start": 1909.7,
                    "end": 1909.98,
                    "word": " open",
                    "probability": 0.8359375
                },
                {
                    "start": 1909.98,
                    "end": 1910.2,
                    "word": " it",
                    "probability": 0.890625
                },
                {
                    "start": 1910.2,
                    "end": 1910.54,
                    "word": " up",
                    "probability": 0.91162109375
                },
                {
                    "start": 1910.54,
                    "end": 1910.82,
                    "word": " for",
                    "probability": 0.89599609375
                },
                {
                    "start": 1910.82,
                    "end": 1911.36,
                    "word": " questions",
                    "probability": 0.90185546875
                },
                {
                    "start": 1911.36,
                    "end": 1911.92,
                    "word": ".",
                    "probability": 0.99072265625
                }
            ]
        },
        {
            "id": 327,
            "seek": 193313,
            "start": 1918.99,
            "end": 1922.67,
            "text": " All right. Thank you for this terrific talk.",
            "tokens": [
                50614,
                1057,
                558,
                13,
                1044,
                291,
                337,
                341,
                20899,
                751,
                13,
                50764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17412725601110374,
            "compression_ratio": 1.4754716981132074,
            "no_speech_prob": 0.0181732177734375,
            "words": [
                {
                    "start": 1918.99,
                    "end": 1919.47,
                    "word": " All",
                    "probability": 0.25146484375
                },
                {
                    "start": 1919.47,
                    "end": 1919.71,
                    "word": " right",
                    "probability": 0.88330078125
                },
                {
                    "start": 1919.71,
                    "end": 1919.91,
                    "word": ".",
                    "probability": 0.5791015625
                },
                {
                    "start": 1919.91,
                    "end": 1920.33,
                    "word": " Thank",
                    "probability": 0.88134765625
                },
                {
                    "start": 1920.33,
                    "end": 1920.61,
                    "word": " you",
                    "probability": 0.8935546875
                },
                {
                    "start": 1920.61,
                    "end": 1921.01,
                    "word": " for",
                    "probability": 0.896484375
                },
                {
                    "start": 1921.01,
                    "end": 1921.43,
                    "word": " this",
                    "probability": 0.8857421875
                },
                {
                    "start": 1921.43,
                    "end": 1921.85,
                    "word": " terrific",
                    "probability": 0.75341796875
                },
                {
                    "start": 1921.85,
                    "end": 1922.37,
                    "word": " talk",
                    "probability": 0.81494140625
                },
                {
                    "start": 1922.37,
                    "end": 1922.67,
                    "word": ".",
                    "probability": 0.96484375
                }
            ]
        },
        {
            "id": 328,
            "seek": 193313,
            "start": 1922.67,
            "end": 1925.03,
            "text": " This was differential privacy, more important than ever",
            "tokens": [
                50814,
                639,
                390,
                15756,
                11427,
                11,
                544,
                1021,
                813,
                1562,
                50914
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17412725601110374,
            "compression_ratio": 1.4754716981132074,
            "no_speech_prob": 0.0181732177734375,
            "words": [
                {
                    "start": 1922.67,
                    "end": 1922.81,
                    "word": " This",
                    "probability": 0.80126953125
                },
                {
                    "start": 1922.81,
                    "end": 1922.97,
                    "word": " was",
                    "probability": 0.87841796875
                },
                {
                    "start": 1922.97,
                    "end": 1923.35,
                    "word": " differential",
                    "probability": 0.369140625
                },
                {
                    "start": 1923.35,
                    "end": 1923.89,
                    "word": " privacy",
                    "probability": 0.83447265625
                },
                {
                    "start": 1923.89,
                    "end": 1924.05,
                    "word": ",",
                    "probability": 0.3671875
                },
                {
                    "start": 1924.05,
                    "end": 1924.19,
                    "word": " more",
                    "probability": 0.86962890625
                },
                {
                    "start": 1924.19,
                    "end": 1924.49,
                    "word": " important",
                    "probability": 0.76953125
                },
                {
                    "start": 1924.49,
                    "end": 1924.75,
                    "word": " than",
                    "probability": 0.88134765625
                },
                {
                    "start": 1924.75,
                    "end": 1925.03,
                    "word": " ever",
                    "probability": 0.81591796875
                }
            ]
        },
        {
            "id": 329,
            "seek": 193313,
            "start": 1925.03,
            "end": 1927.29,
            "text": " in the world of COVID-19.",
            "tokens": [
                50964,
                294,
                264,
                1002,
                295,
                4566,
                12,
                3405,
                13,
                51014
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17412725601110374,
            "compression_ratio": 1.4754716981132074,
            "no_speech_prob": 0.0181732177734375,
            "words": [
                {
                    "start": 1925.03,
                    "end": 1925.53,
                    "word": " in",
                    "probability": 0.85205078125
                },
                {
                    "start": 1925.53,
                    "end": 1925.65,
                    "word": " the",
                    "probability": 0.8134765625
                },
                {
                    "start": 1925.65,
                    "end": 1925.87,
                    "word": " world",
                    "probability": 0.82666015625
                },
                {
                    "start": 1925.87,
                    "end": 1926.03,
                    "word": " of",
                    "probability": 0.90283203125
                },
                {
                    "start": 1926.03,
                    "end": 1926.29,
                    "word": " COVID",
                    "probability": 0.7470703125
                },
                {
                    "start": 1926.29,
                    "end": 1926.73,
                    "word": "-19",
                    "probability": 0.955810546875
                },
                {
                    "start": 1926.73,
                    "end": 1927.29,
                    "word": ".",
                    "probability": 0.96875
                }
            ]
        },
        {
            "id": 330,
            "seek": 193313,
            "start": 1927.67,
            "end": 1931.23,
            "text": " We have Aditi Joshi, Miguel Guevara, and Royce Wilson here",
            "tokens": [
                51064,
                492,
                362,
                1999,
                8707,
                9785,
                72,
                11,
                29150,
                44847,
                85,
                2419,
                11,
                293,
                8751,
                384,
                15388,
                510,
                51214
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17412725601110374,
            "compression_ratio": 1.4754716981132074,
            "no_speech_prob": 0.0181732177734375,
            "words": [
                {
                    "start": 1927.67,
                    "end": 1928.19,
                    "word": " We",
                    "probability": 0.87255859375
                },
                {
                    "start": 1928.19,
                    "end": 1928.43,
                    "word": " have",
                    "probability": 0.876953125
                },
                {
                    "start": 1928.43,
                    "end": 1928.67,
                    "word": " Aditi",
                    "probability": 0.781982421875
                },
                {
                    "start": 1928.67,
                    "end": 1929.15,
                    "word": " Joshi",
                    "probability": 0.92529296875
                },
                {
                    "start": 1929.15,
                    "end": 1929.25,
                    "word": ",",
                    "probability": 0.9208984375
                },
                {
                    "start": 1929.25,
                    "end": 1929.49,
                    "word": " Miguel",
                    "probability": 0.79052734375
                },
                {
                    "start": 1929.49,
                    "end": 1929.91,
                    "word": " Guevara",
                    "probability": 0.939453125
                },
                {
                    "start": 1929.91,
                    "end": 1930.13,
                    "word": ",",
                    "probability": 0.495849609375
                },
                {
                    "start": 1930.13,
                    "end": 1930.19,
                    "word": " and",
                    "probability": 0.89306640625
                },
                {
                    "start": 1930.19,
                    "end": 1930.53,
                    "word": " Royce",
                    "probability": 0.901611328125
                },
                {
                    "start": 1930.53,
                    "end": 1930.75,
                    "word": " Wilson",
                    "probability": 0.85400390625
                },
                {
                    "start": 1930.75,
                    "end": 1931.23,
                    "word": " here",
                    "probability": 0.80224609375
                }
            ]
        },
        {
            "id": 331,
            "seek": 193313,
            "start": 1931.23,
            "end": 1935.17,
            "text": " to chat with us and answer some questions.",
            "tokens": [
                51264,
                281,
                5081,
                365,
                505,
                293,
                1867,
                512,
                1651,
                13,
                51414
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17412725601110374,
            "compression_ratio": 1.4754716981132074,
            "no_speech_prob": 0.0181732177734375,
            "words": [
                {
                    "start": 1931.23,
                    "end": 1931.43,
                    "word": " to",
                    "probability": 0.9013671875
                },
                {
                    "start": 1931.43,
                    "end": 1931.79,
                    "word": " chat",
                    "probability": 0.8076171875
                },
                {
                    "start": 1931.79,
                    "end": 1932.01,
                    "word": " with",
                    "probability": 0.81982421875
                },
                {
                    "start": 1932.01,
                    "end": 1932.41,
                    "word": " us",
                    "probability": 0.9091796875
                },
                {
                    "start": 1932.41,
                    "end": 1933.77,
                    "word": " and",
                    "probability": 0.81201171875
                },
                {
                    "start": 1933.77,
                    "end": 1934.17,
                    "word": " answer",
                    "probability": 0.88232421875
                },
                {
                    "start": 1934.17,
                    "end": 1934.39,
                    "word": " some",
                    "probability": 0.80712890625
                },
                {
                    "start": 1934.39,
                    "end": 1934.99,
                    "word": " questions",
                    "probability": 0.90185546875
                },
                {
                    "start": 1934.99,
                    "end": 1935.17,
                    "word": ".",
                    "probability": 0.984375
                }
            ]
        },
        {
            "id": 332,
            "seek": 193313,
            "start": 1935.17,
            "end": 1938.39,
            "text": " First, I think, Aditi, you wanted to introduce the team.",
            "tokens": [
                51464,
                2386,
                11,
                286,
                519,
                11,
                1999,
                8707,
                11,
                291,
                1415,
                281,
                5366,
                264,
                1469,
                13,
                51564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17412725601110374,
            "compression_ratio": 1.4754716981132074,
            "no_speech_prob": 0.0181732177734375,
            "words": [
                {
                    "start": 1935.17,
                    "end": 1935.31,
                    "word": " First",
                    "probability": 0.89306640625
                },
                {
                    "start": 1935.31,
                    "end": 1935.43,
                    "word": ",",
                    "probability": 0.87841796875
                },
                {
                    "start": 1935.43,
                    "end": 1935.51,
                    "word": " I",
                    "probability": 0.9765625
                },
                {
                    "start": 1935.51,
                    "end": 1935.73,
                    "word": " think",
                    "probability": 0.8037109375
                },
                {
                    "start": 1935.73,
                    "end": 1936.03,
                    "word": ",",
                    "probability": 0.463134765625
                },
                {
                    "start": 1936.03,
                    "end": 1936.19,
                    "word": " Aditi",
                    "probability": 0.976318359375
                },
                {
                    "start": 1936.19,
                    "end": 1936.27,
                    "word": ",",
                    "probability": 0.94189453125
                },
                {
                    "start": 1936.27,
                    "end": 1936.39,
                    "word": " you",
                    "probability": 0.8896484375
                },
                {
                    "start": 1936.39,
                    "end": 1936.61,
                    "word": " wanted",
                    "probability": 0.78662109375
                },
                {
                    "start": 1936.61,
                    "end": 1936.91,
                    "word": " to",
                    "probability": 0.90380859375
                },
                {
                    "start": 1936.91,
                    "end": 1937.27,
                    "word": " introduce",
                    "probability": 0.68408203125
                },
                {
                    "start": 1937.27,
                    "end": 1937.55,
                    "word": " the",
                    "probability": 0.81396484375
                },
                {
                    "start": 1937.55,
                    "end": 1937.83,
                    "word": " team",
                    "probability": 0.9189453125
                },
                {
                    "start": 1937.83,
                    "end": 1938.39,
                    "word": ".",
                    "probability": 0.927734375
                }
            ]
        },
        {
            "id": 333,
            "seek": 193313,
            "start": 1972.29,
            "end": 1975.75,
            "text": " Hold on one moment. I think we're having an audio problem.",
            "tokens": [
                51614,
                6962,
                322,
                472,
                1623,
                13,
                286,
                519,
                321,
                434,
                1419,
                364,
                6278,
                1154,
                13,
                51814
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17412725601110374,
            "compression_ratio": 1.4754716981132074,
            "no_speech_prob": 0.0181732177734375,
            "words": [
                {
                    "start": 1972.29,
                    "end": 1972.63,
                    "word": " Hold",
                    "probability": 0.74658203125
                },
                {
                    "start": 1972.63,
                    "end": 1972.77,
                    "word": " on",
                    "probability": 0.89990234375
                },
                {
                    "start": 1972.77,
                    "end": 1973.13,
                    "word": " one",
                    "probability": 0.6455078125
                },
                {
                    "start": 1973.13,
                    "end": 1973.41,
                    "word": " moment",
                    "probability": 0.90966796875
                },
                {
                    "start": 1973.41,
                    "end": 1973.47,
                    "word": ".",
                    "probability": 0.85400390625
                },
                {
                    "start": 1973.47,
                    "end": 1973.57,
                    "word": " I",
                    "probability": 0.9990234375
                },
                {
                    "start": 1973.57,
                    "end": 1973.73,
                    "word": " think",
                    "probability": 0.81396484375
                },
                {
                    "start": 1973.73,
                    "end": 1973.87,
                    "word": " we're",
                    "probability": 0.927490234375
                },
                {
                    "start": 1973.87,
                    "end": 1974.13,
                    "word": " having",
                    "probability": 0.79736328125
                },
                {
                    "start": 1974.13,
                    "end": 1974.75,
                    "word": " an",
                    "probability": 0.80712890625
                },
                {
                    "start": 1974.75,
                    "end": 1975.09,
                    "word": " audio",
                    "probability": 0.91357421875
                },
                {
                    "start": 1975.09,
                    "end": 1975.51,
                    "word": " problem",
                    "probability": 0.77197265625
                },
                {
                    "start": 1975.51,
                    "end": 1975.75,
                    "word": ".",
                    "probability": 0.76318359375
                }
            ]
        },
        {
            "id": 334,
            "seek": 195993,
            "start": 1978.67,
            "end": 1981.03,
            "text": " We'll wait for a cue.",
            "tokens": [
                50364,
                492,
                603,
                1699,
                337,
                257,
                22656,
                13,
                50564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.22232143538338797,
            "compression_ratio": 1.5811965811965811,
            "no_speech_prob": 0.00042128562927246094,
            "words": [
                {
                    "start": 1978.67,
                    "end": 1979.15,
                    "word": " We'll",
                    "probability": 0.68603515625
                },
                {
                    "start": 1979.15,
                    "end": 1979.63,
                    "word": " wait",
                    "probability": 0.806640625
                },
                {
                    "start": 1979.63,
                    "end": 1979.89,
                    "word": " for",
                    "probability": 0.89990234375
                },
                {
                    "start": 1980.27,
                    "end": 1980.55,
                    "word": " a cue",
                    "probability": 0.346435546875
                },
                {
                    "start": 1980.55,
                    "end": 1981.03,
                    "word": ".",
                    "probability": 0.7333984375
                }
            ]
        },
        {
            "id": 335,
            "seek": 195993,
            "start": 1981.07,
            "end": 1984.03,
            "text": " In the meantime, thanks, everybody, for your patience.",
            "tokens": [
                50614,
                682,
                264,
                14991,
                11,
                3231,
                11,
                2201,
                11,
                337,
                428,
                14826,
                13,
                50714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.22232143538338797,
            "compression_ratio": 1.5811965811965811,
            "no_speech_prob": 0.00042128562927246094,
            "words": [
                {
                    "start": 1981.07,
                    "end": 1981.43,
                    "word": " In",
                    "probability": 0.87255859375
                },
                {
                    "start": 1981.43,
                    "end": 1981.47,
                    "word": " the",
                    "probability": 0.80810546875
                },
                {
                    "start": 1981.47,
                    "end": 1981.87,
                    "word": " meantime",
                    "probability": 0.82666015625
                },
                {
                    "start": 1981.87,
                    "end": 1982.33,
                    "word": ",",
                    "probability": 0.931640625
                },
                {
                    "start": 1982.33,
                    "end": 1982.55,
                    "word": " thanks",
                    "probability": 0.72802734375
                },
                {
                    "start": 1982.55,
                    "end": 1982.61,
                    "word": ",",
                    "probability": 0.314208984375
                },
                {
                    "start": 1982.61,
                    "end": 1982.83,
                    "word": " everybody",
                    "probability": 0.7548828125
                },
                {
                    "start": 1982.83,
                    "end": 1982.99,
                    "word": ",",
                    "probability": 0.8505859375
                },
                {
                    "start": 1982.99,
                    "end": 1983.07,
                    "word": " for",
                    "probability": 0.89892578125
                },
                {
                    "start": 1983.07,
                    "end": 1983.25,
                    "word": " your",
                    "probability": 0.8115234375
                },
                {
                    "start": 1983.25,
                    "end": 1983.59,
                    "word": " patience",
                    "probability": 0.74462890625
                },
                {
                    "start": 1983.59,
                    "end": 1984.03,
                    "word": ".",
                    "probability": 0.98583984375
                }
            ]
        },
        {
            "id": 336,
            "seek": 195993,
            "start": 1984.03,
            "end": 1986.21,
            "text": " This is our glitched episode.",
            "tokens": [
                50764,
                639,
                307,
                527,
                23552,
                292,
                3500,
                13,
                50864
            ],
            "temperature": 0.0,
            "avg_logprob": -0.22232143538338797,
            "compression_ratio": 1.5811965811965811,
            "no_speech_prob": 0.00042128562927246094,
            "words": [
                {
                    "start": 1984.03,
                    "end": 1984.43,
                    "word": " This",
                    "probability": 0.80615234375
                },
                {
                    "start": 1984.43,
                    "end": 1984.67,
                    "word": " is",
                    "probability": 0.90673828125
                },
                {
                    "start": 1984.67,
                    "end": 1984.89,
                    "word": " our",
                    "probability": 0.8134765625
                },
                {
                    "start": 1984.89,
                    "end": 1985.31,
                    "word": " glitched",
                    "probability": 0.837158203125
                },
                {
                    "start": 1985.31,
                    "end": 1985.73,
                    "word": " episode",
                    "probability": 0.7353515625
                },
                {
                    "start": 1985.73,
                    "end": 1986.21,
                    "word": ".",
                    "probability": 0.92431640625
                }
            ]
        },
        {
            "id": 337,
            "seek": 195993,
            "start": 2001.64,
            "end": 2004.26,
            "text": " And just as a reminder, you can submit any questions you have",
            "tokens": [
                50914,
                400,
                445,
                382,
                257,
                13548,
                11,
                291,
                393,
                10315,
                604,
                1651,
                291,
                362,
                51014
            ],
            "temperature": 0.0,
            "avg_logprob": -0.22232143538338797,
            "compression_ratio": 1.5811965811965811,
            "no_speech_prob": 0.00042128562927246094,
            "words": [
                {
                    "start": 2001.64,
                    "end": 2002.04,
                    "word": " And",
                    "probability": 0.8173828125
                },
                {
                    "start": 2002.04,
                    "end": 2002.2,
                    "word": " just",
                    "probability": 0.796875
                },
                {
                    "start": 2002.2,
                    "end": 2002.34,
                    "word": " as",
                    "probability": 0.8955078125
                },
                {
                    "start": 2002.46,
                    "end": 2002.72,
                    "word": " a reminder",
                    "probability": 0.69287109375
                },
                {
                    "start": 2002.72,
                    "end": 2002.86,
                    "word": ",",
                    "probability": 0.974609375
                },
                {
                    "start": 2002.86,
                    "end": 2002.96,
                    "word": " you",
                    "probability": 0.8876953125
                },
                {
                    "start": 2002.96,
                    "end": 2003.1,
                    "word": " can",
                    "probability": 0.80224609375
                },
                {
                    "start": 2003.1,
                    "end": 2003.38,
                    "word": " submit",
                    "probability": 0.818359375
                },
                {
                    "start": 2003.38,
                    "end": 2003.6,
                    "word": " any",
                    "probability": 0.83154296875
                },
                {
                    "start": 2003.6,
                    "end": 2003.94,
                    "word": " questions",
                    "probability": 0.90966796875
                },
                {
                    "start": 2003.94,
                    "end": 2004.1,
                    "word": " you",
                    "probability": 0.8896484375
                },
                {
                    "start": 2004.1,
                    "end": 2004.26,
                    "word": " have",
                    "probability": 0.88134765625
                }
            ]
        },
        {
            "id": 338,
            "seek": 195993,
            "start": 2004.26,
            "end": 2007.28,
            "text": " for the team into the Q&A channel on the Discord.",
            "tokens": [
                51064,
                337,
                264,
                1469,
                666,
                264,
                1249,
                5,
                32,
                2269,
                322,
                264,
                32623,
                13,
                51164
            ],
            "temperature": 0.0,
            "avg_logprob": -0.22232143538338797,
            "compression_ratio": 1.5811965811965811,
            "no_speech_prob": 0.00042128562927246094,
            "words": [
                {
                    "start": 2004.26,
                    "end": 2004.36,
                    "word": " for",
                    "probability": 0.88671875
                },
                {
                    "start": 2004.36,
                    "end": 2004.5,
                    "word": " the",
                    "probability": 0.810546875
                },
                {
                    "start": 2004.5,
                    "end": 2004.8,
                    "word": " team",
                    "probability": 0.92333984375
                },
                {
                    "start": 2004.8,
                    "end": 2005.08,
                    "word": " into",
                    "probability": 0.7216796875
                },
                {
                    "start": 2005.08,
                    "end": 2005.32,
                    "word": " the",
                    "probability": 0.8076171875
                },
                {
                    "start": 2005.32,
                    "end": 2005.46,
                    "word": " Q",
                    "probability": 0.994140625
                },
                {
                    "start": 2005.46,
                    "end": 2005.64,
                    "word": "&A",
                    "probability": 0.956787109375
                },
                {
                    "start": 2005.64,
                    "end": 2005.98,
                    "word": " channel",
                    "probability": 0.79052734375
                },
                {
                    "start": 2005.98,
                    "end": 2006.3,
                    "word": " on",
                    "probability": 0.89990234375
                },
                {
                    "start": 2006.3,
                    "end": 2006.48,
                    "word": " the",
                    "probability": 0.80029296875
                },
                {
                    "start": 2006.48,
                    "end": 2006.8,
                    "word": " Discord",
                    "probability": 0.72900390625
                },
                {
                    "start": 2006.8,
                    "end": 2007.28,
                    "word": ".",
                    "probability": 0.9423828125
                }
            ]
        },
        {
            "id": 339,
            "seek": 195993,
            "start": 2007.48,
            "end": 2012.18,
            "text": " You can find the link there at the link from the Twitch stream.",
            "tokens": [
                51214,
                509,
                393,
                915,
                264,
                2113,
                456,
                412,
                264,
                2113,
                490,
                264,
                22222,
                4309,
                13,
                51414
            ],
            "temperature": 0.0,
            "avg_logprob": -0.22232143538338797,
            "compression_ratio": 1.5811965811965811,
            "no_speech_prob": 0.00042128562927246094,
            "words": [
                {
                    "start": 2007.48,
                    "end": 2007.66,
                    "word": " You",
                    "probability": 0.80322265625
                },
                {
                    "start": 2007.66,
                    "end": 2007.78,
                    "word": " can",
                    "probability": 0.88134765625
                },
                {
                    "start": 2007.78,
                    "end": 2008.0,
                    "word": " find",
                    "probability": 0.83203125
                },
                {
                    "start": 2008.0,
                    "end": 2008.2,
                    "word": " the",
                    "probability": 0.81005859375
                },
                {
                    "start": 2008.2,
                    "end": 2008.5,
                    "word": " link",
                    "probability": 0.861328125
                },
                {
                    "start": 2008.5,
                    "end": 2009.72,
                    "word": " there",
                    "probability": 0.2357177734375
                },
                {
                    "start": 2009.72,
                    "end": 2009.86,
                    "word": " at",
                    "probability": 0.54443359375
                },
                {
                    "start": 2009.86,
                    "end": 2010.16,
                    "word": " the",
                    "probability": 0.80517578125
                },
                {
                    "start": 2010.16,
                    "end": 2010.62,
                    "word": " link",
                    "probability": 0.39697265625
                },
                {
                    "start": 2010.62,
                    "end": 2010.94,
                    "word": " from",
                    "probability": 0.8310546875
                },
                {
                    "start": 2010.94,
                    "end": 2011.18,
                    "word": " the",
                    "probability": 0.8056640625
                },
                {
                    "start": 2011.18,
                    "end": 2011.38,
                    "word": " Twitch",
                    "probability": 0.93994140625
                },
                {
                    "start": 2011.38,
                    "end": 2011.82,
                    "word": " stream",
                    "probability": 0.87939453125
                },
                {
                    "start": 2011.82,
                    "end": 2012.18,
                    "word": ".",
                    "probability": 0.955078125
                }
            ]
        },
        {
            "id": 340,
            "seek": 195993,
            "start": 2012.18,
            "end": 2015.16,
            "text": " You can also go to our website. It contains links to Twitch.",
            "tokens": [
                51464,
                509,
                393,
                611,
                352,
                281,
                527,
                3144,
                13,
                467,
                8306,
                6123,
                281,
                22222,
                13,
                51564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.22232143538338797,
            "compression_ratio": 1.5811965811965811,
            "no_speech_prob": 0.00042128562927246094,
            "words": [
                {
                    "start": 2012.18,
                    "end": 2012.34,
                    "word": " You",
                    "probability": 0.87060546875
                },
                {
                    "start": 2012.34,
                    "end": 2012.46,
                    "word": " can",
                    "probability": 0.89892578125
                },
                {
                    "start": 2012.46,
                    "end": 2012.64,
                    "word": " also",
                    "probability": 0.82373046875
                },
                {
                    "start": 2012.64,
                    "end": 2012.76,
                    "word": " go",
                    "probability": 0.88720703125
                },
                {
                    "start": 2012.76,
                    "end": 2012.86,
                    "word": " to",
                    "probability": 0.908203125
                },
                {
                    "start": 2012.86,
                    "end": 2013.0,
                    "word": " our",
                    "probability": 0.806640625
                },
                {
                    "start": 2013.0,
                    "end": 2013.46,
                    "word": " website",
                    "probability": 0.841796875
                },
                {
                    "start": 2013.46,
                    "end": 2013.6,
                    "word": ".",
                    "probability": 0.54931640625
                },
                {
                    "start": 2013.6,
                    "end": 2013.64,
                    "word": " It",
                    "probability": 0.7392578125
                },
                {
                    "start": 2013.64,
                    "end": 2013.88,
                    "word": " contains",
                    "probability": 0.857421875
                },
                {
                    "start": 2013.88,
                    "end": 2014.18,
                    "word": " links",
                    "probability": 0.806640625
                },
                {
                    "start": 2014.18,
                    "end": 2014.36,
                    "word": " to",
                    "probability": 0.8974609375
                },
                {
                    "start": 2014.36,
                    "end": 2014.68,
                    "word": " Twitch",
                    "probability": 0.72265625
                },
                {
                    "start": 2014.68,
                    "end": 2015.16,
                    "word": ".",
                    "probability": 0.12890625
                }
            ]
        },
        {
            "id": 341,
            "seek": 195993,
            "start": 2015.8,
            "end": 2017.46,
            "text": " Excuse me, to our Discord.",
            "tokens": [
                51614,
                11359,
                385,
                11,
                281,
                527,
                32623,
                13,
                51714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.22232143538338797,
            "compression_ratio": 1.5811965811965811,
            "no_speech_prob": 0.00042128562927246094,
            "words": [
                {
                    "start": 2015.8,
                    "end": 2016.28,
                    "word": " Excuse",
                    "probability": 0.7822265625
                },
                {
                    "start": 2016.28,
                    "end": 2016.44,
                    "word": " me",
                    "probability": 0.90576171875
                },
                {
                    "start": 2016.44,
                    "end": 2016.44,
                    "word": ",",
                    "probability": 0.87109375
                },
                {
                    "start": 2016.44,
                    "end": 2016.58,
                    "word": " to",
                    "probability": 0.90234375
                },
                {
                    "start": 2016.58,
                    "end": 2016.74,
                    "word": " our",
                    "probability": 0.8134765625
                },
                {
                    "start": 2016.74,
                    "end": 2016.98,
                    "word": " Discord",
                    "probability": 0.8388671875
                },
                {
                    "start": 2016.98,
                    "end": 2017.46,
                    "word": ".",
                    "probability": 0.99609375
                }
            ]
        },
        {
            "id": 342,
            "seek": 198871,
            "start": 2017.46,
            "end": 2045.49,
            "text": " I have her try speaking.",
            "tokens": [
                50414,
                286,
                362,
                720,
                853,
                4124,
                13,
                50514
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1728750014305115,
            "compression_ratio": 1.5913978494623655,
            "no_speech_prob": 0.006473541259765625,
            "words": [
                {
                    "start": 2017.46,
                    "end": 2017.82,
                    "word": " I",
                    "probability": 0.143798828125
                },
                {
                    "start": 2042.99,
                    "end": 2044.29,
                    "word": " have",
                    "probability": 0.69580078125
                },
                {
                    "start": 2044.29,
                    "end": 2044.43,
                    "word": " her",
                    "probability": 0.6103515625
                },
                {
                    "start": 2044.43,
                    "end": 2044.65,
                    "word": " try",
                    "probability": 0.873046875
                },
                {
                    "start": 2044.65,
                    "end": 2045.09,
                    "word": " speaking",
                    "probability": 0.78369140625
                },
                {
                    "start": 2045.09,
                    "end": 2045.49,
                    "word": ".",
                    "probability": 0.8076171875
                }
            ]
        },
        {
            "id": 343,
            "seek": 198871,
            "start": 2085.73,
            "end": 2089.79,
            "text": " So that it was as useful and practical to the audience as possible.",
            "tokens": [
                50564,
                407,
                300,
                309,
                390,
                382,
                4420,
                293,
                8496,
                281,
                264,
                4034,
                382,
                1944,
                13,
                50764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1728750014305115,
            "compression_ratio": 1.5913978494623655,
            "no_speech_prob": 0.006473541259765625,
            "words": [
                {
                    "start": 2085.73,
                    "end": 2086.13,
                    "word": " So",
                    "probability": 0.55224609375
                },
                {
                    "start": 2086.13,
                    "end": 2086.29,
                    "word": " that",
                    "probability": 0.82958984375
                },
                {
                    "start": 2086.29,
                    "end": 2086.37,
                    "word": " it",
                    "probability": 0.8701171875
                },
                {
                    "start": 2086.37,
                    "end": 2086.47,
                    "word": " was",
                    "probability": 0.8701171875
                },
                {
                    "start": 2086.47,
                    "end": 2086.63,
                    "word": " as",
                    "probability": 0.87744140625
                },
                {
                    "start": 2086.63,
                    "end": 2086.97,
                    "word": " useful",
                    "probability": 0.77685546875
                },
                {
                    "start": 2086.97,
                    "end": 2087.23,
                    "word": " and",
                    "probability": 0.89013671875
                },
                {
                    "start": 2087.23,
                    "end": 2087.61,
                    "word": " practical",
                    "probability": 0.82666015625
                },
                {
                    "start": 2087.61,
                    "end": 2088.37,
                    "word": " to",
                    "probability": 0.8896484375
                },
                {
                    "start": 2088.37,
                    "end": 2088.51,
                    "word": " the",
                    "probability": 0.81103515625
                },
                {
                    "start": 2088.51,
                    "end": 2088.79,
                    "word": " audience",
                    "probability": 0.79541015625
                },
                {
                    "start": 2088.79,
                    "end": 2088.99,
                    "word": " as",
                    "probability": 0.90234375
                },
                {
                    "start": 2088.99,
                    "end": 2089.39,
                    "word": " possible",
                    "probability": 0.84912109375
                },
                {
                    "start": 2089.39,
                    "end": 2089.79,
                    "word": ".",
                    "probability": 0.93017578125
                }
            ]
        },
        {
            "id": 344,
            "seek": 198871,
            "start": 2089.87,
            "end": 2092.21,
            "text": " And with that, I'm going to turn it over to Miguel and Royce",
            "tokens": [
                50814,
                400,
                365,
                300,
                11,
                286,
                478,
                516,
                281,
                1261,
                309,
                670,
                281,
                29150,
                293,
                8751,
                384,
                50914
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1728750014305115,
            "compression_ratio": 1.5913978494623655,
            "no_speech_prob": 0.006473541259765625,
            "words": [
                {
                    "start": 2089.87,
                    "end": 2089.95,
                    "word": " And",
                    "probability": 0.83837890625
                },
                {
                    "start": 2089.95,
                    "end": 2090.07,
                    "word": " with",
                    "probability": 0.7998046875
                },
                {
                    "start": 2090.07,
                    "end": 2090.21,
                    "word": " that",
                    "probability": 0.87841796875
                },
                {
                    "start": 2090.21,
                    "end": 2090.29,
                    "word": ",",
                    "probability": 0.59521484375
                },
                {
                    "start": 2090.29,
                    "end": 2090.33,
                    "word": " I'm",
                    "probability": 0.980712890625
                },
                {
                    "start": 2090.33,
                    "end": 2090.39,
                    "word": " going",
                    "probability": 0.69482421875
                },
                {
                    "start": 2090.39,
                    "end": 2090.47,
                    "word": " to",
                    "probability": 0.90234375
                },
                {
                    "start": 2090.47,
                    "end": 2090.63,
                    "word": " turn",
                    "probability": 0.81201171875
                },
                {
                    "start": 2090.63,
                    "end": 2090.73,
                    "word": " it",
                    "probability": 0.90576171875
                },
                {
                    "start": 2090.73,
                    "end": 2090.97,
                    "word": " over",
                    "probability": 0.88232421875
                },
                {
                    "start": 2090.97,
                    "end": 2091.29,
                    "word": " to",
                    "probability": 0.90673828125
                },
                {
                    "start": 2091.29,
                    "end": 2091.63,
                    "word": " Miguel",
                    "probability": 0.84423828125
                },
                {
                    "start": 2091.63,
                    "end": 2091.87,
                    "word": " and",
                    "probability": 0.89794921875
                },
                {
                    "start": 2091.87,
                    "end": 2092.21,
                    "word": " Royce",
                    "probability": 0.83349609375
                }
            ]
        },
        {
            "id": 345,
            "seek": 198871,
            "start": 2092.21,
            "end": 2093.51,
            "text": " to introduce themselves.",
            "tokens": [
                50964,
                281,
                5366,
                2969,
                13,
                51014
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1728750014305115,
            "compression_ratio": 1.5913978494623655,
            "no_speech_prob": 0.006473541259765625,
            "words": [
                {
                    "start": 2092.21,
                    "end": 2092.37,
                    "word": " to",
                    "probability": 0.87451171875
                },
                {
                    "start": 2092.37,
                    "end": 2092.69,
                    "word": " introduce",
                    "probability": 0.712890625
                },
                {
                    "start": 2092.69,
                    "end": 2093.11,
                    "word": " themselves",
                    "probability": 0.73876953125
                },
                {
                    "start": 2093.11,
                    "end": 2093.51,
                    "word": ".",
                    "probability": 0.99267578125
                }
            ]
        },
        {
            "id": 346,
            "seek": 198871,
            "start": 2093.99,
            "end": 2097.43,
            "text": " All right. One moment. Let me check with our team behind the scenes.",
            "tokens": [
                51064,
                1057,
                558,
                13,
                1485,
                1623,
                13,
                961,
                385,
                1520,
                365,
                527,
                1469,
                2261,
                264,
                8026,
                13,
                51164
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1728750014305115,
            "compression_ratio": 1.5913978494623655,
            "no_speech_prob": 0.006473541259765625,
            "words": [
                {
                    "start": 2093.99,
                    "end": 2094.39,
                    "word": " All",
                    "probability": 0.79345703125
                },
                {
                    "start": 2094.39,
                    "end": 2094.51,
                    "word": " right",
                    "probability": 0.880859375
                },
                {
                    "start": 2094.51,
                    "end": 2094.55,
                    "word": ".",
                    "probability": 0.54150390625
                },
                {
                    "start": 2094.55,
                    "end": 2094.79,
                    "word": " One",
                    "probability": 0.892578125
                },
                {
                    "start": 2094.79,
                    "end": 2095.07,
                    "word": " moment",
                    "probability": 0.91748046875
                },
                {
                    "start": 2095.07,
                    "end": 2095.47,
                    "word": ".",
                    "probability": 0.91455078125
                },
                {
                    "start": 2095.47,
                    "end": 2095.77,
                    "word": " Let",
                    "probability": 0.88720703125
                },
                {
                    "start": 2095.77,
                    "end": 2095.87,
                    "word": " me",
                    "probability": 0.9091796875
                },
                {
                    "start": 2095.87,
                    "end": 2096.15,
                    "word": " check",
                    "probability": 0.837890625
                },
                {
                    "start": 2096.15,
                    "end": 2096.37,
                    "word": " with",
                    "probability": 0.82275390625
                },
                {
                    "start": 2096.37,
                    "end": 2096.55,
                    "word": " our",
                    "probability": 0.80810546875
                },
                {
                    "start": 2096.55,
                    "end": 2096.75,
                    "word": " team",
                    "probability": 0.92822265625
                },
                {
                    "start": 2096.75,
                    "end": 2096.97,
                    "word": " behind",
                    "probability": 0.83154296875
                },
                {
                    "start": 2096.97,
                    "end": 2097.15,
                    "word": " the",
                    "probability": 0.80859375
                },
                {
                    "start": 2097.15,
                    "end": 2097.35,
                    "word": " scenes",
                    "probability": 0.7763671875
                },
                {
                    "start": 2097.35,
                    "end": 2097.43,
                    "word": ".",
                    "probability": 0.92333984375
                }
            ]
        },
        {
            "id": 347,
            "seek": 198871,
            "start": 2097.43,
            "end": 2099.47,
            "text": " Do we have audio right now for our speakers?",
            "tokens": [
                51214,
                1144,
                321,
                362,
                6278,
                558,
                586,
                337,
                527,
                9518,
                30,
                51314
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1728750014305115,
            "compression_ratio": 1.5913978494623655,
            "no_speech_prob": 0.006473541259765625,
            "words": [
                {
                    "start": 2097.43,
                    "end": 2097.49,
                    "word": " Do",
                    "probability": 0.91943359375
                },
                {
                    "start": 2097.49,
                    "end": 2097.61,
                    "word": " we",
                    "probability": 0.90625
                },
                {
                    "start": 2097.61,
                    "end": 2097.77,
                    "word": " have",
                    "probability": 0.87939453125
                },
                {
                    "start": 2097.77,
                    "end": 2098.09,
                    "word": " audio",
                    "probability": 0.91455078125
                },
                {
                    "start": 2098.09,
                    "end": 2098.33,
                    "word": " right",
                    "probability": 0.86962890625
                },
                {
                    "start": 2098.33,
                    "end": 2098.49,
                    "word": " now",
                    "probability": 0.89501953125
                },
                {
                    "start": 2098.49,
                    "end": 2098.65,
                    "word": " for",
                    "probability": 0.8955078125
                },
                {
                    "start": 2098.65,
                    "end": 2098.81,
                    "word": " our",
                    "probability": 0.82568359375
                },
                {
                    "start": 2098.81,
                    "end": 2099.07,
                    "word": " speakers",
                    "probability": 0.79443359375
                },
                {
                    "start": 2099.07,
                    "end": 2099.47,
                    "word": "?",
                    "probability": 0.99365234375
                }
            ]
        },
        {
            "id": 348,
            "seek": 198871,
            "start": 2102.86,
            "end": 2105.1,
            "text": " Okay, we do. Great. Fantastic. Wonderful.",
            "tokens": [
                51364,
                1033,
                11,
                321,
                360,
                13,
                3769,
                13,
                21320,
                13,
                22768,
                13,
                51464
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1728750014305115,
            "compression_ratio": 1.5913978494623655,
            "no_speech_prob": 0.006473541259765625,
            "words": [
                {
                    "start": 2102.86,
                    "end": 2103.26,
                    "word": " Okay",
                    "probability": 0.385498046875
                },
                {
                    "start": 2103.26,
                    "end": 2103.32,
                    "word": ",",
                    "probability": 0.59716796875
                },
                {
                    "start": 2103.32,
                    "end": 2103.44,
                    "word": " we",
                    "probability": 0.9033203125
                },
                {
                    "start": 2103.44,
                    "end": 2103.6,
                    "word": " do",
                    "probability": 0.90185546875
                },
                {
                    "start": 2103.6,
                    "end": 2103.7,
                    "word": ".",
                    "probability": 0.9296875
                },
                {
                    "start": 2103.7,
                    "end": 2103.84,
                    "word": " Great",
                    "probability": 0.84912109375
                },
                {
                    "start": 2103.84,
                    "end": 2103.94,
                    "word": ".",
                    "probability": 0.74365234375
                },
                {
                    "start": 2103.96,
                    "end": 2104.36,
                    "word": " Fantastic",
                    "probability": 0.9130859375
                },
                {
                    "start": 2104.36,
                    "end": 2104.58,
                    "word": ".",
                    "probability": 0.900390625
                },
                {
                    "start": 2104.58,
                    "end": 2104.84,
                    "word": " Wonderful",
                    "probability": 0.73583984375
                },
                {
                    "start": 2104.84,
                    "end": 2105.1,
                    "word": ".",
                    "probability": 0.99169921875
                }
            ]
        },
        {
            "id": 349,
            "seek": 198871,
            "start": 2105.1,
            "end": 2107.56,
            "text": " Thanks for that intro, Aditi. Miguel, take it away.",
            "tokens": [
                51514,
                2561,
                337,
                300,
                12897,
                11,
                1999,
                8707,
                13,
                29150,
                11,
                747,
                309,
                1314,
                13,
                51614
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1728750014305115,
            "compression_ratio": 1.5913978494623655,
            "no_speech_prob": 0.006473541259765625,
            "words": [
                {
                    "start": 2105.1,
                    "end": 2105.24,
                    "word": " Thanks",
                    "probability": 0.78369140625
                },
                {
                    "start": 2105.24,
                    "end": 2105.36,
                    "word": " for",
                    "probability": 0.89990234375
                },
                {
                    "start": 2105.36,
                    "end": 2105.5,
                    "word": " that",
                    "probability": 0.59521484375
                },
                {
                    "start": 2105.5,
                    "end": 2105.8,
                    "word": " intro",
                    "probability": 0.81640625
                },
                {
                    "start": 2105.8,
                    "end": 2105.9,
                    "word": ",",
                    "probability": 0.87158203125
                },
                {
                    "start": 2105.9,
                    "end": 2106.1,
                    "word": " Aditi",
                    "probability": 0.90087890625
                },
                {
                    "start": 2106.1,
                    "end": 2106.28,
                    "word": ".",
                    "probability": 0.98095703125
                },
                {
                    "start": 2106.28,
                    "end": 2106.56,
                    "word": " Miguel",
                    "probability": 0.8193359375
                },
                {
                    "start": 2106.56,
                    "end": 2106.66,
                    "word": ",",
                    "probability": 0.974609375
                },
                {
                    "start": 2106.66,
                    "end": 2106.92,
                    "word": " take",
                    "probability": 0.8154296875
                },
                {
                    "start": 2106.92,
                    "end": 2106.98,
                    "word": " it",
                    "probability": 0.90625
                },
                {
                    "start": 2106.98,
                    "end": 2107.16,
                    "word": " away",
                    "probability": 0.81201171875
                },
                {
                    "start": 2107.16,
                    "end": 2107.56,
                    "word": ".",
                    "probability": 0.99462890625
                }
            ]
        },
        {
            "id": 350,
            "seek": 198871,
            "start": 2108.18,
            "end": 2111.78,
            "text": " Great. Thank you, everyone, for coming today to our talk.",
            "tokens": [
                51664,
                3769,
                13,
                1044,
                291,
                11,
                1518,
                11,
                337,
                1348,
                965,
                281,
                527,
                751,
                13,
                51764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1728750014305115,
            "compression_ratio": 1.5913978494623655,
            "no_speech_prob": 0.006473541259765625,
            "words": [
                {
                    "start": 2108.18,
                    "end": 2108.58,
                    "word": " Great",
                    "probability": 0.85888671875
                },
                {
                    "start": 2108.58,
                    "end": 2108.98,
                    "word": ".",
                    "probability": 0.9599609375
                },
                {
                    "start": 2109.34,
                    "end": 2109.74,
                    "word": " Thank",
                    "probability": 0.88330078125
                },
                {
                    "start": 2109.74,
                    "end": 2109.84,
                    "word": " you",
                    "probability": 0.8896484375
                },
                {
                    "start": 2109.84,
                    "end": 2109.9,
                    "word": ",",
                    "probability": 0.59521484375
                },
                {
                    "start": 2109.9,
                    "end": 2110.08,
                    "word": " everyone",
                    "probability": 0.8837890625
                },
                {
                    "start": 2110.08,
                    "end": 2110.3,
                    "word": ",",
                    "probability": 0.9169921875
                },
                {
                    "start": 2110.3,
                    "end": 2110.48,
                    "word": " for",
                    "probability": 0.90185546875
                },
                {
                    "start": 2110.48,
                    "end": 2110.86,
                    "word": " coming",
                    "probability": 0.82373046875
                },
                {
                    "start": 2110.86,
                    "end": 2111.1,
                    "word": " today",
                    "probability": 0.76904296875
                },
                {
                    "start": 2111.1,
                    "end": 2111.24,
                    "word": " to",
                    "probability": 0.89794921875
                },
                {
                    "start": 2111.24,
                    "end": 2111.38,
                    "word": " our",
                    "probability": 0.814453125
                },
                {
                    "start": 2111.38,
                    "end": 2111.68,
                    "word": " talk",
                    "probability": 0.8173828125
                },
                {
                    "start": 2111.68,
                    "end": 2111.78,
                    "word": ".",
                    "probability": 0.9873046875
                }
            ]
        },
        {
            "id": 351,
            "seek": 201467,
            "start": 2111.78,
            "end": 2116.16,
            "text": " I'm Miguel. I'm a product manager working on Google's privacy team.",
            "tokens": [
                50364,
                286,
                478,
                29150,
                13,
                286,
                478,
                257,
                1674,
                6598,
                1364,
                322,
                3329,
                311,
                11427,
                1469,
                13,
                50514
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20475260758151612,
            "compression_ratio": 1.6340425531914893,
            "no_speech_prob": 0.01031494140625,
            "words": [
                {
                    "start": 2111.78,
                    "end": 2112.02,
                    "word": " I'm",
                    "probability": 0.636962890625
                },
                {
                    "start": 2112.02,
                    "end": 2112.36,
                    "word": " Miguel",
                    "probability": 0.6630859375
                },
                {
                    "start": 2112.36,
                    "end": 2112.66,
                    "word": ".",
                    "probability": 0.5849609375
                },
                {
                    "start": 2112.66,
                    "end": 2112.76,
                    "word": " I'm",
                    "probability": 0.9873046875
                },
                {
                    "start": 2112.9,
                    "end": 2113.16,
                    "word": " a product",
                    "probability": 0.6708984375
                },
                {
                    "start": 2113.16,
                    "end": 2113.52,
                    "word": " manager",
                    "probability": 0.8720703125
                },
                {
                    "start": 2113.52,
                    "end": 2113.96,
                    "word": " working",
                    "probability": 0.7373046875
                },
                {
                    "start": 2113.96,
                    "end": 2114.42,
                    "word": " on",
                    "probability": 0.9130859375
                },
                {
                    "start": 2114.42,
                    "end": 2114.86,
                    "word": " Google's",
                    "probability": 0.927490234375
                },
                {
                    "start": 2114.86,
                    "end": 2115.16,
                    "word": " privacy",
                    "probability": 0.63671875
                },
                {
                    "start": 2115.16,
                    "end": 2115.56,
                    "word": " team",
                    "probability": 0.90576171875
                },
                {
                    "start": 2115.56,
                    "end": 2116.16,
                    "word": ".",
                    "probability": 0.9775390625
                }
            ]
        },
        {
            "id": 352,
            "seek": 201467,
            "start": 2116.3,
            "end": 2119.5,
            "text": " Basically, I work with Royce and other engineers on the team",
            "tokens": [
                50564,
                8537,
                11,
                286,
                589,
                365,
                8751,
                384,
                293,
                661,
                11955,
                322,
                264,
                1469,
                50714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20475260758151612,
            "compression_ratio": 1.6340425531914893,
            "no_speech_prob": 0.01031494140625,
            "words": [
                {
                    "start": 2116.3,
                    "end": 2116.9,
                    "word": " Basically",
                    "probability": 0.87646484375
                },
                {
                    "start": 2116.9,
                    "end": 2117.2,
                    "word": ",",
                    "probability": 0.74658203125
                },
                {
                    "start": 2117.2,
                    "end": 2117.3,
                    "word": " I",
                    "probability": 0.9990234375
                },
                {
                    "start": 2117.3,
                    "end": 2117.5,
                    "word": " work",
                    "probability": 0.7890625
                },
                {
                    "start": 2117.5,
                    "end": 2117.66,
                    "word": " with",
                    "probability": 0.8115234375
                },
                {
                    "start": 2117.66,
                    "end": 2117.98,
                    "word": " Royce",
                    "probability": 0.7685546875
                },
                {
                    "start": 2117.98,
                    "end": 2118.14,
                    "word": " and",
                    "probability": 0.8759765625
                },
                {
                    "start": 2118.14,
                    "end": 2118.4,
                    "word": " other",
                    "probability": 0.78515625
                },
                {
                    "start": 2118.4,
                    "end": 2118.8,
                    "word": " engineers",
                    "probability": 0.716796875
                },
                {
                    "start": 2118.8,
                    "end": 2119.0,
                    "word": " on",
                    "probability": 0.8994140625
                },
                {
                    "start": 2119.0,
                    "end": 2119.12,
                    "word": " the",
                    "probability": 0.7275390625
                },
                {
                    "start": 2119.12,
                    "end": 2119.5,
                    "word": " team",
                    "probability": 0.9189453125
                }
            ]
        },
        {
            "id": 353,
            "seek": 201467,
            "start": 2119.5,
            "end": 2123.24,
            "text": " on developing our internal differential private infrastructure",
            "tokens": [
                50764,
                322,
                6416,
                527,
                6920,
                15756,
                4551,
                6896,
                50914
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20475260758151612,
            "compression_ratio": 1.6340425531914893,
            "no_speech_prob": 0.01031494140625,
            "words": [
                {
                    "start": 2119.5,
                    "end": 2119.78,
                    "word": " on",
                    "probability": 0.869140625
                },
                {
                    "start": 2119.78,
                    "end": 2120.22,
                    "word": " developing",
                    "probability": 0.65087890625
                },
                {
                    "start": 2120.22,
                    "end": 2120.66,
                    "word": " our",
                    "probability": 0.80810546875
                },
                {
                    "start": 2120.66,
                    "end": 2121.18,
                    "word": " internal",
                    "probability": 0.69384765625
                },
                {
                    "start": 2121.18,
                    "end": 2122.12,
                    "word": " differential",
                    "probability": 0.43115234375
                },
                {
                    "start": 2122.12,
                    "end": 2122.56,
                    "word": " private",
                    "probability": 0.76904296875
                },
                {
                    "start": 2122.56,
                    "end": 2123.24,
                    "word": " infrastructure",
                    "probability": 0.85791015625
                }
            ]
        },
        {
            "id": 354,
            "seek": 201467,
            "start": 2123.54,
            "end": 2126.04,
            "text": " that we recently open sourced.",
            "tokens": [
                50964,
                300,
                321,
                3938,
                1269,
                11006,
                1232,
                13,
                51064
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20475260758151612,
            "compression_ratio": 1.6340425531914893,
            "no_speech_prob": 0.01031494140625,
            "words": [
                {
                    "start": 2123.54,
                    "end": 2123.84,
                    "word": " that",
                    "probability": 0.82373046875
                },
                {
                    "start": 2123.84,
                    "end": 2123.94,
                    "word": " we",
                    "probability": 0.90380859375
                },
                {
                    "start": 2123.94,
                    "end": 2124.6,
                    "word": " recently",
                    "probability": 0.71728515625
                },
                {
                    "start": 2124.6,
                    "end": 2125.04,
                    "word": " open",
                    "probability": 0.44775390625
                },
                {
                    "start": 2125.04,
                    "end": 2125.44,
                    "word": " sourced",
                    "probability": 0.80517578125
                },
                {
                    "start": 2125.44,
                    "end": 2126.04,
                    "word": ".",
                    "probability": 0.99072265625
                }
            ]
        },
        {
            "id": 355,
            "seek": 201467,
            "start": 2129.54,
            "end": 2133.0,
            "text": " And then I am Royce. I'm a sophomore engineer at Google.",
            "tokens": [
                51114,
                400,
                550,
                286,
                669,
                8751,
                384,
                13,
                286,
                478,
                257,
                35798,
                11403,
                412,
                3329,
                13,
                51264
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20475260758151612,
            "compression_ratio": 1.6340425531914893,
            "no_speech_prob": 0.01031494140625,
            "words": [
                {
                    "start": 2129.54,
                    "end": 2130.14,
                    "word": " And",
                    "probability": 0.490478515625
                },
                {
                    "start": 2130.14,
                    "end": 2130.28,
                    "word": " then",
                    "probability": 0.70556640625
                },
                {
                    "start": 2130.28,
                    "end": 2130.46,
                    "word": " I",
                    "probability": 0.76220703125
                },
                {
                    "start": 2130.46,
                    "end": 2130.66,
                    "word": " am",
                    "probability": 0.8681640625
                },
                {
                    "start": 2130.66,
                    "end": 2131.02,
                    "word": " Royce",
                    "probability": 0.927734375
                },
                {
                    "start": 2131.02,
                    "end": 2131.22,
                    "word": ".",
                    "probability": 0.92578125
                },
                {
                    "start": 2131.22,
                    "end": 2131.54,
                    "word": " I'm",
                    "probability": 0.94775390625
                },
                {
                    "start": 2131.64,
                    "end": 2131.88,
                    "word": " a sophomore",
                    "probability": 0.603515625
                },
                {
                    "start": 2131.88,
                    "end": 2132.18,
                    "word": " engineer",
                    "probability": 0.7197265625
                },
                {
                    "start": 2132.18,
                    "end": 2132.4,
                    "word": " at",
                    "probability": 0.8955078125
                },
                {
                    "start": 2132.4,
                    "end": 2132.6,
                    "word": " Google",
                    "probability": 0.884765625
                },
                {
                    "start": 2132.6,
                    "end": 2133.0,
                    "word": ".",
                    "probability": 0.3720703125
                }
            ]
        },
        {
            "id": 356,
            "seek": 201467,
            "start": 2133.0,
            "end": 2135.64,
            "text": " I work on privacy infrastructure.",
            "tokens": [
                51314,
                286,
                589,
                322,
                11427,
                6896,
                13,
                51414
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20475260758151612,
            "compression_ratio": 1.6340425531914893,
            "no_speech_prob": 0.01031494140625,
            "words": [
                {
                    "start": 2133.0,
                    "end": 2133.3,
                    "word": " I",
                    "probability": 0.66943359375
                },
                {
                    "start": 2133.3,
                    "end": 2133.88,
                    "word": " work",
                    "probability": 0.77197265625
                },
                {
                    "start": 2133.88,
                    "end": 2134.14,
                    "word": " on",
                    "probability": 0.9091796875
                },
                {
                    "start": 2134.14,
                    "end": 2134.48,
                    "word": " privacy",
                    "probability": 0.8076171875
                },
                {
                    "start": 2134.48,
                    "end": 2135.04,
                    "word": " infrastructure",
                    "probability": 0.861328125
                },
                {
                    "start": 2135.04,
                    "end": 2135.64,
                    "word": ".",
                    "probability": 0.8642578125
                }
            ]
        },
        {
            "id": 357,
            "seek": 201467,
            "start": 2135.96,
            "end": 2140.36,
            "text": " I was also the lead author behind the paper that the talk is based on.",
            "tokens": [
                51464,
                286,
                390,
                611,
                264,
                1477,
                3793,
                2261,
                264,
                3035,
                300,
                264,
                751,
                307,
                2361,
                322,
                13,
                51614
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20475260758151612,
            "compression_ratio": 1.6340425531914893,
            "no_speech_prob": 0.01031494140625,
            "words": [
                {
                    "start": 2135.96,
                    "end": 2136.3,
                    "word": " I",
                    "probability": 0.97314453125
                },
                {
                    "start": 2136.3,
                    "end": 2136.38,
                    "word": " was",
                    "probability": 0.88671875
                },
                {
                    "start": 2136.38,
                    "end": 2136.66,
                    "word": " also",
                    "probability": 0.822265625
                },
                {
                    "start": 2136.66,
                    "end": 2137.22,
                    "word": " the",
                    "probability": 0.81591796875
                },
                {
                    "start": 2137.22,
                    "end": 2137.48,
                    "word": " lead",
                    "probability": 0.85009765625
                },
                {
                    "start": 2137.48,
                    "end": 2137.66,
                    "word": " author",
                    "probability": 0.7880859375
                },
                {
                    "start": 2137.66,
                    "end": 2137.96,
                    "word": " behind",
                    "probability": 0.84765625
                },
                {
                    "start": 2137.96,
                    "end": 2138.38,
                    "word": " the",
                    "probability": 0.810546875
                },
                {
                    "start": 2138.38,
                    "end": 2138.68,
                    "word": " paper",
                    "probability": 0.8037109375
                },
                {
                    "start": 2138.68,
                    "end": 2138.88,
                    "word": " that",
                    "probability": 0.88134765625
                },
                {
                    "start": 2138.88,
                    "end": 2139.04,
                    "word": " the",
                    "probability": 0.8017578125
                },
                {
                    "start": 2139.04,
                    "end": 2139.24,
                    "word": " talk",
                    "probability": 0.736328125
                },
                {
                    "start": 2139.24,
                    "end": 2139.36,
                    "word": " is",
                    "probability": 0.381103515625
                },
                {
                    "start": 2139.36,
                    "end": 2139.54,
                    "word": " based",
                    "probability": 0.8046875
                },
                {
                    "start": 2139.54,
                    "end": 2139.76,
                    "word": " on",
                    "probability": 0.90625
                },
                {
                    "start": 2139.76,
                    "end": 2140.36,
                    "word": ".",
                    "probability": 0.92724609375
                }
            ]
        },
        {
            "id": 358,
            "seek": 204267,
            "start": 2145.11,
            "end": 2148.17,
            "text": " Terrific. So we have a couple of questions.",
            "tokens": [
                50414,
                36591,
                1089,
                13,
                407,
                321,
                362,
                257,
                1916,
                295,
                1651,
                13,
                50564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17187499898975178,
            "compression_ratio": 1.8764478764478765,
            "no_speech_prob": 0.011138916015625,
            "words": [
                {
                    "start": 2145.11,
                    "end": 2145.55,
                    "word": " Terrific",
                    "probability": 0.5435791015625
                },
                {
                    "start": 2145.55,
                    "end": 2145.99,
                    "word": ".",
                    "probability": 0.7060546875
                },
                {
                    "start": 2145.99,
                    "end": 2146.41,
                    "word": " So",
                    "probability": 0.7998046875
                },
                {
                    "start": 2146.41,
                    "end": 2146.53,
                    "word": " we",
                    "probability": 0.81298828125
                },
                {
                    "start": 2146.53,
                    "end": 2146.67,
                    "word": " have",
                    "probability": 0.875
                },
                {
                    "start": 2146.81,
                    "end": 2147.01,
                    "word": " a couple",
                    "probability": 0.78857421875
                },
                {
                    "start": 2147.01,
                    "end": 2147.21,
                    "word": " of",
                    "probability": 0.8974609375
                },
                {
                    "start": 2147.21,
                    "end": 2147.73,
                    "word": " questions",
                    "probability": 0.8984375
                },
                {
                    "start": 2147.73,
                    "end": 2148.17,
                    "word": ".",
                    "probability": 0.91748046875
                }
            ]
        },
        {
            "id": 359,
            "seek": 204267,
            "start": 2148.77,
            "end": 2150.97,
            "text": " I'll start off with the question that I have,",
            "tokens": [
                50614,
                286,
                603,
                722,
                766,
                365,
                264,
                1168,
                300,
                286,
                362,
                11,
                50714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17187499898975178,
            "compression_ratio": 1.8764478764478765,
            "no_speech_prob": 0.011138916015625,
            "words": [
                {
                    "start": 2148.77,
                    "end": 2149.21,
                    "word": " I'll",
                    "probability": 0.95166015625
                },
                {
                    "start": 2149.21,
                    "end": 2149.47,
                    "word": " start",
                    "probability": 0.86767578125
                },
                {
                    "start": 2149.47,
                    "end": 2149.71,
                    "word": " off",
                    "probability": 0.7763671875
                },
                {
                    "start": 2149.71,
                    "end": 2149.93,
                    "word": " with",
                    "probability": 0.82373046875
                },
                {
                    "start": 2149.93,
                    "end": 2150.09,
                    "word": " the",
                    "probability": 0.8037109375
                },
                {
                    "start": 2150.09,
                    "end": 2150.35,
                    "word": " question",
                    "probability": 0.78564453125
                },
                {
                    "start": 2150.35,
                    "end": 2150.49,
                    "word": " that",
                    "probability": 0.78076171875
                },
                {
                    "start": 2150.49,
                    "end": 2150.61,
                    "word": " I",
                    "probability": 0.99609375
                },
                {
                    "start": 2150.61,
                    "end": 2150.87,
                    "word": " have",
                    "probability": 0.8779296875
                },
                {
                    "start": 2150.87,
                    "end": 2150.97,
                    "word": ",",
                    "probability": 0.64208984375
                }
            ]
        },
        {
            "id": 360,
            "seek": 204267,
            "start": 2150.97,
            "end": 2153.77,
            "text": " just taking the liberty and privilege to do so.",
            "tokens": [
                50764,
                445,
                1940,
                264,
                22849,
                293,
                12122,
                281,
                360,
                370,
                13,
                50864
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17187499898975178,
            "compression_ratio": 1.8764478764478765,
            "no_speech_prob": 0.011138916015625,
            "words": [
                {
                    "start": 2150.97,
                    "end": 2151.19,
                    "word": " just",
                    "probability": 0.888671875
                },
                {
                    "start": 2151.19,
                    "end": 2151.65,
                    "word": " taking",
                    "probability": 0.81689453125
                },
                {
                    "start": 2151.65,
                    "end": 2152.07,
                    "word": " the",
                    "probability": 0.81884765625
                },
                {
                    "start": 2152.07,
                    "end": 2152.31,
                    "word": " liberty",
                    "probability": 0.8173828125
                },
                {
                    "start": 2152.31,
                    "end": 2152.51,
                    "word": " and",
                    "probability": 0.89013671875
                },
                {
                    "start": 2152.51,
                    "end": 2152.73,
                    "word": " privilege",
                    "probability": 0.6640625
                },
                {
                    "start": 2152.73,
                    "end": 2152.93,
                    "word": " to",
                    "probability": 0.90625
                },
                {
                    "start": 2152.93,
                    "end": 2153.05,
                    "word": " do",
                    "probability": 0.8984375
                },
                {
                    "start": 2153.05,
                    "end": 2153.33,
                    "word": " so",
                    "probability": 0.8974609375
                },
                {
                    "start": 2153.33,
                    "end": 2153.77,
                    "word": ".",
                    "probability": 0.96826171875
                }
            ]
        },
        {
            "id": 361,
            "seek": 204267,
            "start": 2154.19,
            "end": 2156.79,
            "text": " Have you encountered any misunderstandings in your work",
            "tokens": [
                50914,
                3560,
                291,
                20381,
                604,
                35736,
                1109,
                294,
                428,
                589,
                51014
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17187499898975178,
            "compression_ratio": 1.8764478764478765,
            "no_speech_prob": 0.011138916015625,
            "words": [
                {
                    "start": 2154.19,
                    "end": 2154.47,
                    "word": " Have",
                    "probability": 0.83544921875
                },
                {
                    "start": 2154.47,
                    "end": 2154.61,
                    "word": " you",
                    "probability": 0.89111328125
                },
                {
                    "start": 2154.61,
                    "end": 2154.97,
                    "word": " encountered",
                    "probability": 0.61669921875
                },
                {
                    "start": 2154.97,
                    "end": 2155.27,
                    "word": " any",
                    "probability": 0.7880859375
                },
                {
                    "start": 2155.27,
                    "end": 2156.27,
                    "word": " misunderstandings",
                    "probability": 0.84326171875
                },
                {
                    "start": 2156.27,
                    "end": 2156.45,
                    "word": " in",
                    "probability": 0.90185546875
                },
                {
                    "start": 2156.45,
                    "end": 2156.59,
                    "word": " your",
                    "probability": 0.80224609375
                },
                {
                    "start": 2156.59,
                    "end": 2156.79,
                    "word": " work",
                    "probability": 0.8125
                }
            ]
        },
        {
            "id": 362,
            "seek": 204267,
            "start": 2156.79,
            "end": 2158.81,
            "text": " about how Google uses differential privacy?",
            "tokens": [
                51064,
                466,
                577,
                3329,
                4960,
                15756,
                11427,
                30,
                51114
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17187499898975178,
            "compression_ratio": 1.8764478764478765,
            "no_speech_prob": 0.011138916015625,
            "words": [
                {
                    "start": 2156.79,
                    "end": 2156.97,
                    "word": " about",
                    "probability": 0.81298828125
                },
                {
                    "start": 2156.97,
                    "end": 2157.15,
                    "word": " how",
                    "probability": 0.90478515625
                },
                {
                    "start": 2157.15,
                    "end": 2157.37,
                    "word": " Google",
                    "probability": 0.8857421875
                },
                {
                    "start": 2157.37,
                    "end": 2157.71,
                    "word": " uses",
                    "probability": 0.8330078125
                },
                {
                    "start": 2157.71,
                    "end": 2158.13,
                    "word": " differential",
                    "probability": 0.69189453125
                },
                {
                    "start": 2158.13,
                    "end": 2158.55,
                    "word": " privacy",
                    "probability": 0.85693359375
                },
                {
                    "start": 2158.55,
                    "end": 2158.81,
                    "word": "?",
                    "probability": 0.461669921875
                }
            ]
        },
        {
            "id": 363,
            "seek": 204267,
            "start": 2158.81,
            "end": 2160.27,
            "text": " And what have you found to be most effective",
            "tokens": [
                51164,
                400,
                437,
                362,
                291,
                1352,
                281,
                312,
                881,
                4942,
                51214
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17187499898975178,
            "compression_ratio": 1.8764478764478765,
            "no_speech_prob": 0.011138916015625,
            "words": [
                {
                    "start": 2158.81,
                    "end": 2158.91,
                    "word": " And",
                    "probability": 0.88623046875
                },
                {
                    "start": 2158.91,
                    "end": 2159.03,
                    "word": " what",
                    "probability": 0.892578125
                },
                {
                    "start": 2159.03,
                    "end": 2159.15,
                    "word": " have",
                    "probability": 0.876953125
                },
                {
                    "start": 2159.15,
                    "end": 2159.25,
                    "word": " you",
                    "probability": 0.8876953125
                },
                {
                    "start": 2159.25,
                    "end": 2159.47,
                    "word": " found",
                    "probability": 0.888671875
                },
                {
                    "start": 2159.47,
                    "end": 2159.57,
                    "word": " to",
                    "probability": 0.900390625
                },
                {
                    "start": 2159.57,
                    "end": 2159.69,
                    "word": " be",
                    "probability": 0.90673828125
                },
                {
                    "start": 2159.69,
                    "end": 2159.87,
                    "word": " most",
                    "probability": 0.81494140625
                },
                {
                    "start": 2159.87,
                    "end": 2160.27,
                    "word": " effective",
                    "probability": 0.7490234375
                }
            ]
        },
        {
            "id": 364,
            "seek": 204267,
            "start": 2160.27,
            "end": 2162.91,
            "text": " in combating any of those misunderstandings?",
            "tokens": [
                51264,
                294,
                2512,
                990,
                604,
                295,
                729,
                35736,
                1109,
                30,
                51314
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17187499898975178,
            "compression_ratio": 1.8764478764478765,
            "no_speech_prob": 0.011138916015625,
            "words": [
                {
                    "start": 2160.27,
                    "end": 2160.81,
                    "word": " in",
                    "probability": 0.904296875
                },
                {
                    "start": 2160.81,
                    "end": 2161.17,
                    "word": " combating",
                    "probability": 0.833251953125
                },
                {
                    "start": 2161.17,
                    "end": 2161.39,
                    "word": " any",
                    "probability": 0.78662109375
                },
                {
                    "start": 2161.39,
                    "end": 2161.51,
                    "word": " of",
                    "probability": 0.9091796875
                },
                {
                    "start": 2161.51,
                    "end": 2161.65,
                    "word": " those",
                    "probability": 0.85498046875
                },
                {
                    "start": 2161.65,
                    "end": 2162.47,
                    "word": " misunderstandings",
                    "probability": 0.90283203125
                },
                {
                    "start": 2162.47,
                    "end": 2162.91,
                    "word": "?",
                    "probability": 0.98828125
                }
            ]
        },
        {
            "id": 365,
            "seek": 204267,
            "start": 2164.33,
            "end": 2165.73,
            "text": " That's an interesting question.",
            "tokens": [
                51364,
                663,
                311,
                364,
                1880,
                1168,
                13,
                51464
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17187499898975178,
            "compression_ratio": 1.8764478764478765,
            "no_speech_prob": 0.011138916015625,
            "words": [
                {
                    "start": 2164.33,
                    "end": 2164.77,
                    "word": " That's",
                    "probability": 0.873779296875
                },
                {
                    "start": 2164.77,
                    "end": 2164.89,
                    "word": " an",
                    "probability": 0.908203125
                },
                {
                    "start": 2164.89,
                    "end": 2165.11,
                    "word": " interesting",
                    "probability": 0.64794921875
                },
                {
                    "start": 2165.11,
                    "end": 2165.51,
                    "word": " question",
                    "probability": 0.8046875
                },
                {
                    "start": 2165.51,
                    "end": 2165.73,
                    "word": ".",
                    "probability": 0.98779296875
                }
            ]
        },
        {
            "id": 366,
            "seek": 204267,
            "start": 2165.91,
            "end": 2169.81,
            "text": " I wouldn't say it's misunderstandings on how Google uses differential privacy.",
            "tokens": [
                51514,
                286,
                2759,
                380,
                584,
                309,
                311,
                35736,
                1109,
                322,
                577,
                3329,
                4960,
                15756,
                11427,
                13,
                51664
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17187499898975178,
            "compression_ratio": 1.8764478764478765,
            "no_speech_prob": 0.011138916015625,
            "words": [
                {
                    "start": 2165.91,
                    "end": 2166.35,
                    "word": " I",
                    "probability": 0.99609375
                },
                {
                    "start": 2166.35,
                    "end": 2166.63,
                    "word": " wouldn't",
                    "probability": 0.906494140625
                },
                {
                    "start": 2166.63,
                    "end": 2166.87,
                    "word": " say",
                    "probability": 0.89501953125
                },
                {
                    "start": 2166.87,
                    "end": 2167.21,
                    "word": " it's",
                    "probability": 0.952392578125
                },
                {
                    "start": 2167.21,
                    "end": 2167.93,
                    "word": " misunderstandings",
                    "probability": 0.8408203125
                },
                {
                    "start": 2167.93,
                    "end": 2168.07,
                    "word": " on",
                    "probability": 0.7451171875
                },
                {
                    "start": 2168.07,
                    "end": 2168.27,
                    "word": " how",
                    "probability": 0.8994140625
                },
                {
                    "start": 2168.27,
                    "end": 2168.61,
                    "word": " Google",
                    "probability": 0.8974609375
                },
                {
                    "start": 2168.61,
                    "end": 2168.95,
                    "word": " uses",
                    "probability": 0.81494140625
                },
                {
                    "start": 2168.95,
                    "end": 2169.37,
                    "word": " differential",
                    "probability": 0.7021484375
                },
                {
                    "start": 2169.37,
                    "end": 2169.81,
                    "word": " privacy",
                    "probability": 0.82421875
                },
                {
                    "start": 2169.81,
                    "end": 2169.81,
                    "word": ".",
                    "probability": 0.916015625
                }
            ]
        },
        {
            "id": 367,
            "seek": 204267,
            "start": 2169.81,
            "end": 2171.55,
            "text": " I would say it's just your own misunderstanding",
            "tokens": [
                51714,
                286,
                576,
                584,
                309,
                311,
                445,
                428,
                1065,
                29227,
                51764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17187499898975178,
            "compression_ratio": 1.8764478764478765,
            "no_speech_prob": 0.011138916015625,
            "words": [
                {
                    "start": 2169.81,
                    "end": 2170.11,
                    "word": " I",
                    "probability": 0.998046875
                },
                {
                    "start": 2170.11,
                    "end": 2170.25,
                    "word": " would",
                    "probability": 0.84619140625
                },
                {
                    "start": 2170.25,
                    "end": 2170.41,
                    "word": " say",
                    "probability": 0.89599609375
                },
                {
                    "start": 2170.41,
                    "end": 2170.61,
                    "word": " it's",
                    "probability": 0.94482421875
                },
                {
                    "start": 2170.61,
                    "end": 2170.89,
                    "word": " just",
                    "probability": 0.8662109375
                },
                {
                    "start": 2170.89,
                    "end": 2171.09,
                    "word": " your",
                    "probability": 0.5224609375
                },
                {
                    "start": 2171.09,
                    "end": 2171.31,
                    "word": " own",
                    "probability": 0.794921875
                },
                {
                    "start": 2171.31,
                    "end": 2171.55,
                    "word": " misunderstanding",
                    "probability": 0.8388671875
                }
            ]
        },
        {
            "id": 368,
            "seek": 207011,
            "start": 2171.55,
            "end": 2174.73,
            "text": " about what differential privacy is.",
            "tokens": [
                50364,
                466,
                437,
                15756,
                11427,
                307,
                13,
                50464
            ],
            "temperature": 0.0,
            "avg_logprob": -0.22861043110634516,
            "compression_ratio": 1.7862903225806452,
            "no_speech_prob": 0.00394439697265625,
            "words": [
                {
                    "start": 2171.55,
                    "end": 2172.39,
                    "word": " about",
                    "probability": 0.00745391845703125
                },
                {
                    "start": 2172.39,
                    "end": 2172.93,
                    "word": " what",
                    "probability": 0.78369140625
                },
                {
                    "start": 2172.93,
                    "end": 2173.35,
                    "word": " differential",
                    "probability": 0.626953125
                },
                {
                    "start": 2173.35,
                    "end": 2173.75,
                    "word": " privacy",
                    "probability": 0.83056640625
                },
                {
                    "start": 2173.75,
                    "end": 2174.15,
                    "word": " is",
                    "probability": 0.9072265625
                },
                {
                    "start": 2174.15,
                    "end": 2174.73,
                    "word": ".",
                    "probability": 0.89208984375
                }
            ]
        },
        {
            "id": 369,
            "seek": 207011,
            "start": 2174.73,
            "end": 2178.63,
            "text": " A lot of the times when we launch new features",
            "tokens": [
                50514,
                316,
                688,
                295,
                264,
                1413,
                562,
                321,
                4025,
                777,
                4122,
                50664
            ],
            "temperature": 0.0,
            "avg_logprob": -0.22861043110634516,
            "compression_ratio": 1.7862903225806452,
            "no_speech_prob": 0.00394439697265625,
            "words": [
                {
                    "start": 2174.73,
                    "end": 2175.17,
                    "word": " A",
                    "probability": 0.95703125
                },
                {
                    "start": 2175.17,
                    "end": 2175.29,
                    "word": " lot",
                    "probability": 0.8984375
                },
                {
                    "start": 2175.29,
                    "end": 2175.39,
                    "word": " of",
                    "probability": 0.90087890625
                },
                {
                    "start": 2175.39,
                    "end": 2175.49,
                    "word": " the",
                    "probability": 0.77490234375
                },
                {
                    "start": 2175.49,
                    "end": 2175.69,
                    "word": " times",
                    "probability": 0.7822265625
                },
                {
                    "start": 2175.69,
                    "end": 2176.03,
                    "word": " when",
                    "probability": 0.76513671875
                },
                {
                    "start": 2176.03,
                    "end": 2176.99,
                    "word": " we",
                    "probability": 0.76171875
                },
                {
                    "start": 2176.99,
                    "end": 2177.91,
                    "word": " launch",
                    "probability": 0.8203125
                },
                {
                    "start": 2177.91,
                    "end": 2178.21,
                    "word": " new",
                    "probability": 0.7998046875
                },
                {
                    "start": 2178.21,
                    "end": 2178.63,
                    "word": " features",
                    "probability": 0.748046875
                }
            ]
        },
        {
            "id": 370,
            "seek": 207011,
            "start": 2178.63,
            "end": 2181.11,
            "text": " and when we open source things, people approach to us",
            "tokens": [
                50714,
                293,
                562,
                321,
                1269,
                4009,
                721,
                11,
                561,
                3109,
                281,
                505,
                50814
            ],
            "temperature": 0.0,
            "avg_logprob": -0.22861043110634516,
            "compression_ratio": 1.7862903225806452,
            "no_speech_prob": 0.00394439697265625,
            "words": [
                {
                    "start": 2178.63,
                    "end": 2178.85,
                    "word": " and",
                    "probability": 0.69482421875
                },
                {
                    "start": 2178.85,
                    "end": 2179.03,
                    "word": " when",
                    "probability": 0.791015625
                },
                {
                    "start": 2179.03,
                    "end": 2179.19,
                    "word": " we",
                    "probability": 0.89599609375
                },
                {
                    "start": 2179.19,
                    "end": 2179.43,
                    "word": " open",
                    "probability": 0.79052734375
                },
                {
                    "start": 2179.43,
                    "end": 2179.63,
                    "word": " source",
                    "probability": 0.693359375
                },
                {
                    "start": 2179.63,
                    "end": 2179.95,
                    "word": " things",
                    "probability": 0.76171875
                },
                {
                    "start": 2179.95,
                    "end": 2180.07,
                    "word": ",",
                    "probability": 0.7001953125
                },
                {
                    "start": 2180.07,
                    "end": 2180.19,
                    "word": " people",
                    "probability": 0.92626953125
                },
                {
                    "start": 2180.19,
                    "end": 2180.65,
                    "word": " approach",
                    "probability": 0.54443359375
                },
                {
                    "start": 2180.65,
                    "end": 2180.87,
                    "word": " to",
                    "probability": 0.6181640625
                },
                {
                    "start": 2180.87,
                    "end": 2181.11,
                    "word": " us",
                    "probability": 0.8935546875
                }
            ]
        },
        {
            "id": 371,
            "seek": 207011,
            "start": 2181.11,
            "end": 2183.43,
            "text": " trying to use it. They're very excited",
            "tokens": [
                50864,
                1382,
                281,
                764,
                309,
                13,
                814,
                434,
                588,
                2919,
                50914
            ],
            "temperature": 0.0,
            "avg_logprob": -0.22861043110634516,
            "compression_ratio": 1.7862903225806452,
            "no_speech_prob": 0.00394439697265625,
            "words": [
                {
                    "start": 2181.11,
                    "end": 2181.87,
                    "word": " trying",
                    "probability": 0.54296875
                },
                {
                    "start": 2181.87,
                    "end": 2182.05,
                    "word": " to",
                    "probability": 0.90625
                },
                {
                    "start": 2182.05,
                    "end": 2182.29,
                    "word": " use",
                    "probability": 0.81982421875
                },
                {
                    "start": 2182.29,
                    "end": 2182.41,
                    "word": " it",
                    "probability": 0.89501953125
                },
                {
                    "start": 2182.41,
                    "end": 2182.57,
                    "word": ".",
                    "probability": 0.611328125
                },
                {
                    "start": 2182.57,
                    "end": 2182.81,
                    "word": " They're",
                    "probability": 0.796630859375
                },
                {
                    "start": 2182.81,
                    "end": 2183.01,
                    "word": " very",
                    "probability": 0.72705078125
                },
                {
                    "start": 2183.01,
                    "end": 2183.43,
                    "word": " excited",
                    "probability": 0.8955078125
                }
            ]
        },
        {
            "id": 372,
            "seek": 207011,
            "start": 2183.43,
            "end": 2186.41,
            "text": " because they sort of think that differential privacy",
            "tokens": [
                50964,
                570,
                436,
                1333,
                295,
                519,
                300,
                15756,
                11427,
                51064
            ],
            "temperature": 0.0,
            "avg_logprob": -0.22861043110634516,
            "compression_ratio": 1.7862903225806452,
            "no_speech_prob": 0.00394439697265625,
            "words": [
                {
                    "start": 2183.43,
                    "end": 2184.19,
                    "word": " because",
                    "probability": 0.70361328125
                },
                {
                    "start": 2184.19,
                    "end": 2184.41,
                    "word": " they",
                    "probability": 0.71240234375
                },
                {
                    "start": 2184.41,
                    "end": 2184.61,
                    "word": " sort",
                    "probability": 0.68212890625
                },
                {
                    "start": 2184.61,
                    "end": 2184.73,
                    "word": " of",
                    "probability": 0.912109375
                },
                {
                    "start": 2184.73,
                    "end": 2184.93,
                    "word": " think",
                    "probability": 0.70654296875
                },
                {
                    "start": 2184.93,
                    "end": 2185.45,
                    "word": " that",
                    "probability": 0.75146484375
                },
                {
                    "start": 2185.45,
                    "end": 2185.91,
                    "word": " differential",
                    "probability": 0.751953125
                },
                {
                    "start": 2185.91,
                    "end": 2186.41,
                    "word": " privacy",
                    "probability": 0.82470703125
                }
            ]
        },
        {
            "id": 373,
            "seek": 207011,
            "start": 2186.41,
            "end": 2190.47,
            "text": " is this magic wand that when it touches any sort of data,",
            "tokens": [
                51114,
                307,
                341,
                5585,
                14304,
                300,
                562,
                309,
                17431,
                604,
                1333,
                295,
                1412,
                11,
                51264
            ],
            "temperature": 0.0,
            "avg_logprob": -0.22861043110634516,
            "compression_ratio": 1.7862903225806452,
            "no_speech_prob": 0.00394439697265625,
            "words": [
                {
                    "start": 2186.41,
                    "end": 2186.67,
                    "word": " is",
                    "probability": 0.89599609375
                },
                {
                    "start": 2186.67,
                    "end": 2186.83,
                    "word": " this",
                    "probability": 0.845703125
                },
                {
                    "start": 2186.83,
                    "end": 2187.55,
                    "word": " magic",
                    "probability": 0.41943359375
                },
                {
                    "start": 2187.55,
                    "end": 2187.99,
                    "word": " wand",
                    "probability": 0.7783203125
                },
                {
                    "start": 2187.99,
                    "end": 2188.45,
                    "word": " that",
                    "probability": 0.85205078125
                },
                {
                    "start": 2188.45,
                    "end": 2188.79,
                    "word": " when",
                    "probability": 0.765625
                },
                {
                    "start": 2188.79,
                    "end": 2188.93,
                    "word": " it",
                    "probability": 0.88720703125
                },
                {
                    "start": 2188.93,
                    "end": 2189.25,
                    "word": " touches",
                    "probability": 0.7548828125
                },
                {
                    "start": 2189.25,
                    "end": 2189.69,
                    "word": " any",
                    "probability": 0.794921875
                },
                {
                    "start": 2189.69,
                    "end": 2189.95,
                    "word": " sort",
                    "probability": 0.7919921875
                },
                {
                    "start": 2189.95,
                    "end": 2190.07,
                    "word": " of",
                    "probability": 0.908203125
                },
                {
                    "start": 2190.07,
                    "end": 2190.27,
                    "word": " data",
                    "probability": 0.62646484375
                },
                {
                    "start": 2190.27,
                    "end": 2190.47,
                    "word": ",",
                    "probability": 0.70166015625
                }
            ]
        },
        {
            "id": 374,
            "seek": 207011,
            "start": 2190.47,
            "end": 2193.43,
            "text": " it suddenly transforms it into private data.",
            "tokens": [
                51314,
                309,
                5800,
                35592,
                309,
                666,
                4551,
                1412,
                13,
                51414
            ],
            "temperature": 0.0,
            "avg_logprob": -0.22861043110634516,
            "compression_ratio": 1.7862903225806452,
            "no_speech_prob": 0.00394439697265625,
            "words": [
                {
                    "start": 2190.47,
                    "end": 2190.53,
                    "word": " it",
                    "probability": 0.8583984375
                },
                {
                    "start": 2190.53,
                    "end": 2190.97,
                    "word": " suddenly",
                    "probability": 0.4580078125
                },
                {
                    "start": 2190.97,
                    "end": 2191.47,
                    "word": " transforms",
                    "probability": 0.79541015625
                },
                {
                    "start": 2191.47,
                    "end": 2191.83,
                    "word": " it",
                    "probability": 0.888671875
                },
                {
                    "start": 2191.83,
                    "end": 2192.11,
                    "word": " into",
                    "probability": 0.73046875
                },
                {
                    "start": 2192.11,
                    "end": 2192.51,
                    "word": " private",
                    "probability": 0.82275390625
                },
                {
                    "start": 2192.51,
                    "end": 2192.85,
                    "word": " data",
                    "probability": 0.80078125
                },
                {
                    "start": 2192.85,
                    "end": 2193.43,
                    "word": ".",
                    "probability": 0.88427734375
                }
            ]
        },
        {
            "id": 375,
            "seek": 207011,
            "start": 2193.61,
            "end": 2195.87,
            "text": " And I think that the biggest misunderstanding is",
            "tokens": [
                51464,
                400,
                286,
                519,
                300,
                264,
                3880,
                29227,
                307,
                51564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.22861043110634516,
            "compression_ratio": 1.7862903225806452,
            "no_speech_prob": 0.00394439697265625,
            "words": [
                {
                    "start": 2193.61,
                    "end": 2194.13,
                    "word": " And",
                    "probability": 0.83251953125
                },
                {
                    "start": 2194.13,
                    "end": 2194.27,
                    "word": " I",
                    "probability": 0.994140625
                },
                {
                    "start": 2194.27,
                    "end": 2194.43,
                    "word": " think",
                    "probability": 0.80810546875
                },
                {
                    "start": 2194.43,
                    "end": 2194.57,
                    "word": " that",
                    "probability": 0.8212890625
                },
                {
                    "start": 2194.57,
                    "end": 2194.69,
                    "word": " the",
                    "probability": 0.8115234375
                },
                {
                    "start": 2194.69,
                    "end": 2194.89,
                    "word": " biggest",
                    "probability": 0.9208984375
                },
                {
                    "start": 2194.89,
                    "end": 2195.29,
                    "word": " misunderstanding",
                    "probability": 0.7919921875
                },
                {
                    "start": 2195.29,
                    "end": 2195.87,
                    "word": " is",
                    "probability": 0.90771484375
                }
            ]
        },
        {
            "id": 376,
            "seek": 207011,
            "start": 2195.87,
            "end": 2198.99,
            "text": " people confuse differential privacy a lot with synthetic data.",
            "tokens": [
                51614,
                561,
                28584,
                15756,
                11427,
                257,
                688,
                365,
                23420,
                1412,
                13,
                51714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.22861043110634516,
            "compression_ratio": 1.7862903225806452,
            "no_speech_prob": 0.00394439697265625,
            "words": [
                {
                    "start": 2195.87,
                    "end": 2196.39,
                    "word": " people",
                    "probability": 0.71630859375
                },
                {
                    "start": 2196.39,
                    "end": 2196.81,
                    "word": " confuse",
                    "probability": 0.86083984375
                },
                {
                    "start": 2196.81,
                    "end": 2197.15,
                    "word": " differential",
                    "probability": 0.7626953125
                },
                {
                    "start": 2197.15,
                    "end": 2197.61,
                    "word": " privacy",
                    "probability": 0.81982421875
                },
                {
                    "start": 2197.83,
                    "end": 2197.95,
                    "word": " a lot",
                    "probability": 0.89990234375
                },
                {
                    "start": 2197.95,
                    "end": 2198.19,
                    "word": " with",
                    "probability": 0.8154296875
                },
                {
                    "start": 2198.19,
                    "end": 2198.53,
                    "word": " synthetic",
                    "probability": 0.759765625
                },
                {
                    "start": 2198.53,
                    "end": 2198.91,
                    "word": " data",
                    "probability": 0.80322265625
                },
                {
                    "start": 2198.91,
                    "end": 2198.99,
                    "word": ".",
                    "probability": 0.58203125
                }
            ]
        },
        {
            "id": 377,
            "seek": 209731,
            "start": 2198.99,
            "end": 2201.61,
            "text": " So they believe that differential privacy can automatically",
            "tokens": [
                50364,
                407,
                436,
                1697,
                300,
                15756,
                11427,
                393,
                6772,
                50464
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18261718018246548,
            "compression_ratio": 1.7665369649805447,
            "no_speech_prob": 0.0015001296997070312,
            "words": [
                {
                    "start": 2198.99,
                    "end": 2199.09,
                    "word": " So",
                    "probability": 0.2479248046875
                },
                {
                    "start": 2199.09,
                    "end": 2199.27,
                    "word": " they",
                    "probability": 0.67138671875
                },
                {
                    "start": 2199.27,
                    "end": 2199.99,
                    "word": " believe",
                    "probability": 0.736328125
                },
                {
                    "start": 2199.99,
                    "end": 2200.17,
                    "word": " that",
                    "probability": 0.759765625
                },
                {
                    "start": 2200.17,
                    "end": 2200.43,
                    "word": " differential",
                    "probability": 0.405029296875
                },
                {
                    "start": 2200.43,
                    "end": 2200.71,
                    "word": " privacy",
                    "probability": 0.166748046875
                },
                {
                    "start": 2200.71,
                    "end": 2201.01,
                    "word": " can",
                    "probability": 0.89404296875
                },
                {
                    "start": 2201.01,
                    "end": 2201.61,
                    "word": " automatically",
                    "probability": 0.8154296875
                }
            ]
        },
        {
            "id": 378,
            "seek": 209731,
            "start": 2201.61,
            "end": 2203.79,
            "text": " turn things into synthetic data.",
            "tokens": [
                50514,
                1261,
                721,
                666,
                23420,
                1412,
                13,
                50564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18261718018246548,
            "compression_ratio": 1.7665369649805447,
            "no_speech_prob": 0.0015001296997070312,
            "words": [
                {
                    "start": 2201.61,
                    "end": 2201.95,
                    "word": " turn",
                    "probability": 0.7822265625
                },
                {
                    "start": 2201.95,
                    "end": 2202.25,
                    "word": " things",
                    "probability": 0.70849609375
                },
                {
                    "start": 2202.25,
                    "end": 2202.45,
                    "word": " into",
                    "probability": 0.73193359375
                },
                {
                    "start": 2202.45,
                    "end": 2202.79,
                    "word": " synthetic",
                    "probability": 0.712890625
                },
                {
                    "start": 2202.79,
                    "end": 2203.19,
                    "word": " data",
                    "probability": 0.7939453125
                },
                {
                    "start": 2203.19,
                    "end": 2203.79,
                    "word": ".",
                    "probability": 0.869140625
                }
            ]
        },
        {
            "id": 379,
            "seek": 209731,
            "start": 2204.11,
            "end": 2206.79,
            "text": " And I think that's the biggest misunderstanding.",
            "tokens": [
                50614,
                400,
                286,
                519,
                300,
                311,
                264,
                3880,
                29227,
                13,
                50714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18261718018246548,
            "compression_ratio": 1.7665369649805447,
            "no_speech_prob": 0.0015001296997070312,
            "words": [
                {
                    "start": 2204.11,
                    "end": 2204.71,
                    "word": " And",
                    "probability": 0.80419921875
                },
                {
                    "start": 2204.71,
                    "end": 2205.35,
                    "word": " I",
                    "probability": 0.97216796875
                },
                {
                    "start": 2205.35,
                    "end": 2205.53,
                    "word": " think",
                    "probability": 0.802734375
                },
                {
                    "start": 2205.53,
                    "end": 2205.73,
                    "word": " that's",
                    "probability": 0.938720703125
                },
                {
                    "start": 2205.73,
                    "end": 2205.85,
                    "word": " the",
                    "probability": 0.81640625
                },
                {
                    "start": 2205.85,
                    "end": 2206.01,
                    "word": " biggest",
                    "probability": 0.91455078125
                },
                {
                    "start": 2206.01,
                    "end": 2206.49,
                    "word": " misunderstanding",
                    "probability": 0.78857421875
                },
                {
                    "start": 2206.49,
                    "end": 2206.79,
                    "word": ".",
                    "probability": 0.91796875
                }
            ]
        },
        {
            "id": 380,
            "seek": 209731,
            "start": 2206.79,
            "end": 2209.63,
            "text": " And one thing that's, I think, important to highlight",
            "tokens": [
                50764,
                400,
                472,
                551,
                300,
                311,
                11,
                286,
                519,
                11,
                1021,
                281,
                5078,
                50864
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18261718018246548,
            "compression_ratio": 1.7665369649805447,
            "no_speech_prob": 0.0015001296997070312,
            "words": [
                {
                    "start": 2206.79,
                    "end": 2206.89,
                    "word": " And",
                    "probability": 0.87158203125
                },
                {
                    "start": 2206.89,
                    "end": 2207.25,
                    "word": " one",
                    "probability": 0.81591796875
                },
                {
                    "start": 2207.25,
                    "end": 2207.51,
                    "word": " thing",
                    "probability": 0.8134765625
                },
                {
                    "start": 2207.51,
                    "end": 2207.95,
                    "word": " that's",
                    "probability": 0.89599609375
                },
                {
                    "start": 2207.95,
                    "end": 2207.95,
                    "word": ",",
                    "probability": 0.55908203125
                },
                {
                    "start": 2207.95,
                    "end": 2208.47,
                    "word": " I",
                    "probability": 0.98876953125
                },
                {
                    "start": 2208.47,
                    "end": 2208.63,
                    "word": " think",
                    "probability": 0.8134765625
                },
                {
                    "start": 2208.63,
                    "end": 2208.67,
                    "word": ",",
                    "probability": 0.81689453125
                },
                {
                    "start": 2208.67,
                    "end": 2208.95,
                    "word": " important",
                    "probability": 0.76220703125
                },
                {
                    "start": 2208.95,
                    "end": 2209.17,
                    "word": " to",
                    "probability": 0.90673828125
                },
                {
                    "start": 2209.17,
                    "end": 2209.63,
                    "word": " highlight",
                    "probability": 0.79248046875
                }
            ]
        },
        {
            "id": 381,
            "seek": 209731,
            "start": 2209.63,
            "end": 2212.65,
            "text": " is that our open source offering helps people",
            "tokens": [
                50914,
                307,
                300,
                527,
                1269,
                4009,
                8745,
                3665,
                561,
                51014
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18261718018246548,
            "compression_ratio": 1.7665369649805447,
            "no_speech_prob": 0.0015001296997070312,
            "words": [
                {
                    "start": 2209.63,
                    "end": 2210.09,
                    "word": " is",
                    "probability": 0.888671875
                },
                {
                    "start": 2210.09,
                    "end": 2210.53,
                    "word": " that",
                    "probability": 0.87255859375
                },
                {
                    "start": 2210.53,
                    "end": 2210.99,
                    "word": " our",
                    "probability": 0.470947265625
                },
                {
                    "start": 2210.99,
                    "end": 2211.31,
                    "word": " open",
                    "probability": 0.77587890625
                },
                {
                    "start": 2211.31,
                    "end": 2211.51,
                    "word": " source",
                    "probability": 0.5703125
                },
                {
                    "start": 2211.51,
                    "end": 2211.95,
                    "word": " offering",
                    "probability": 0.728515625
                },
                {
                    "start": 2211.95,
                    "end": 2212.31,
                    "word": " helps",
                    "probability": 0.6064453125
                },
                {
                    "start": 2212.31,
                    "end": 2212.65,
                    "word": " people",
                    "probability": 0.9228515625
                }
            ]
        },
        {
            "id": 382,
            "seek": 209731,
            "start": 2212.65,
            "end": 2215.99,
            "text": " produce statistics in a differential private manner.",
            "tokens": [
                51064,
                5258,
                12523,
                294,
                257,
                15756,
                4551,
                9060,
                13,
                51164
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18261718018246548,
            "compression_ratio": 1.7665369649805447,
            "no_speech_prob": 0.0015001296997070312,
            "words": [
                {
                    "start": 2212.65,
                    "end": 2213.23,
                    "word": " produce",
                    "probability": 0.708984375
                },
                {
                    "start": 2213.23,
                    "end": 2213.87,
                    "word": " statistics",
                    "probability": 0.90673828125
                },
                {
                    "start": 2213.87,
                    "end": 2214.11,
                    "word": " in",
                    "probability": 0.8935546875
                },
                {
                    "start": 2214.21,
                    "end": 2214.55,
                    "word": " a differential",
                    "probability": 0.41455078125
                },
                {
                    "start": 2214.55,
                    "end": 2214.99,
                    "word": " private",
                    "probability": 0.5625
                },
                {
                    "start": 2214.99,
                    "end": 2215.39,
                    "word": " manner",
                    "probability": 0.7978515625
                },
                {
                    "start": 2215.39,
                    "end": 2215.99,
                    "word": ".",
                    "probability": 0.9951171875
                }
            ]
        },
        {
            "id": 383,
            "seek": 209731,
            "start": 2216.01,
            "end": 2219.75,
            "text": " So if any of you has one of those use cases,",
            "tokens": [
                51214,
                407,
                498,
                604,
                295,
                291,
                575,
                472,
                295,
                729,
                764,
                3331,
                11,
                51364
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18261718018246548,
            "compression_ratio": 1.7665369649805447,
            "no_speech_prob": 0.0015001296997070312,
            "words": [
                {
                    "start": 2216.01,
                    "end": 2216.37,
                    "word": " So",
                    "probability": 0.90087890625
                },
                {
                    "start": 2216.37,
                    "end": 2216.97,
                    "word": " if",
                    "probability": 0.83740234375
                },
                {
                    "start": 2216.97,
                    "end": 2217.21,
                    "word": " any",
                    "probability": 0.80908203125
                },
                {
                    "start": 2217.21,
                    "end": 2217.37,
                    "word": " of",
                    "probability": 0.9091796875
                },
                {
                    "start": 2217.37,
                    "end": 2217.53,
                    "word": " you",
                    "probability": 0.89794921875
                },
                {
                    "start": 2217.53,
                    "end": 2217.97,
                    "word": " has",
                    "probability": 0.64111328125
                },
                {
                    "start": 2217.97,
                    "end": 2218.31,
                    "word": " one",
                    "probability": 0.79931640625
                },
                {
                    "start": 2218.31,
                    "end": 2218.43,
                    "word": " of",
                    "probability": 0.9033203125
                },
                {
                    "start": 2218.43,
                    "end": 2218.63,
                    "word": " those",
                    "probability": 0.85595703125
                },
                {
                    "start": 2218.63,
                    "end": 2218.87,
                    "word": " use",
                    "probability": 0.787109375
                },
                {
                    "start": 2218.87,
                    "end": 2219.27,
                    "word": " cases",
                    "probability": 0.8291015625
                },
                {
                    "start": 2219.27,
                    "end": 2219.75,
                    "word": ",",
                    "probability": 0.91650390625
                }
            ]
        },
        {
            "id": 384,
            "seek": 209731,
            "start": 2219.75,
            "end": 2222.45,
            "text": " then I think that the open sourcing offering",
            "tokens": [
                51414,
                550,
                286,
                519,
                300,
                264,
                1269,
                11006,
                2175,
                8745,
                51514
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18261718018246548,
            "compression_ratio": 1.7665369649805447,
            "no_speech_prob": 0.0015001296997070312,
            "words": [
                {
                    "start": 2219.75,
                    "end": 2220.39,
                    "word": " then",
                    "probability": 0.728515625
                },
                {
                    "start": 2220.39,
                    "end": 2220.95,
                    "word": " I",
                    "probability": 0.99072265625
                },
                {
                    "start": 2220.95,
                    "end": 2221.11,
                    "word": " think",
                    "probability": 0.80712890625
                },
                {
                    "start": 2221.11,
                    "end": 2221.25,
                    "word": " that",
                    "probability": 0.86474609375
                },
                {
                    "start": 2221.25,
                    "end": 2221.47,
                    "word": " the",
                    "probability": 0.802734375
                },
                {
                    "start": 2221.47,
                    "end": 2221.75,
                    "word": " open",
                    "probability": 0.79052734375
                },
                {
                    "start": 2221.75,
                    "end": 2222.09,
                    "word": " sourcing",
                    "probability": 0.898681640625
                },
                {
                    "start": 2222.09,
                    "end": 2222.45,
                    "word": " offering",
                    "probability": 0.7578125
                }
            ]
        },
        {
            "id": 385,
            "seek": 209731,
            "start": 2222.45,
            "end": 2224.23,
            "text": " is a good place to start.",
            "tokens": [
                51564,
                307,
                257,
                665,
                1081,
                281,
                722,
                13,
                51614
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18261718018246548,
            "compression_ratio": 1.7665369649805447,
            "no_speech_prob": 0.0015001296997070312,
            "words": [
                {
                    "start": 2222.45,
                    "end": 2222.79,
                    "word": " is",
                    "probability": 0.9013671875
                },
                {
                    "start": 2222.93,
                    "end": 2223.15,
                    "word": " a good",
                    "probability": 0.87939453125
                },
                {
                    "start": 2223.15,
                    "end": 2223.61,
                    "word": " place",
                    "probability": 0.81005859375
                },
                {
                    "start": 2223.61,
                    "end": 2223.75,
                    "word": " to",
                    "probability": 0.90673828125
                },
                {
                    "start": 2223.75,
                    "end": 2224.03,
                    "word": " start",
                    "probability": 0.86865234375
                },
                {
                    "start": 2224.03,
                    "end": 2224.23,
                    "word": ".",
                    "probability": 0.91259765625
                }
            ]
        },
        {
            "id": 386,
            "seek": 209731,
            "start": 2224.23,
            "end": 2226.19,
            "text": " And that's the way that we use it at Google",
            "tokens": [
                51664,
                400,
                300,
                311,
                264,
                636,
                300,
                321,
                764,
                309,
                412,
                3329,
                51714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.18261718018246548,
            "compression_ratio": 1.7665369649805447,
            "no_speech_prob": 0.0015001296997070312,
            "words": [
                {
                    "start": 2224.23,
                    "end": 2224.71,
                    "word": " And",
                    "probability": 0.89501953125
                },
                {
                    "start": 2224.71,
                    "end": 2225.01,
                    "word": " that's",
                    "probability": 0.940673828125
                },
                {
                    "start": 2225.01,
                    "end": 2225.11,
                    "word": " the",
                    "probability": 0.81591796875
                },
                {
                    "start": 2225.11,
                    "end": 2225.27,
                    "word": " way",
                    "probability": 0.89794921875
                },
                {
                    "start": 2225.27,
                    "end": 2225.41,
                    "word": " that",
                    "probability": 0.88037109375
                },
                {
                    "start": 2225.41,
                    "end": 2225.53,
                    "word": " we",
                    "probability": 0.904296875
                },
                {
                    "start": 2225.53,
                    "end": 2225.71,
                    "word": " use",
                    "probability": 0.81103515625
                },
                {
                    "start": 2225.71,
                    "end": 2225.79,
                    "word": " it",
                    "probability": 0.8662109375
                },
                {
                    "start": 2225.79,
                    "end": 2225.95,
                    "word": " at",
                    "probability": 0.89111328125
                },
                {
                    "start": 2225.95,
                    "end": 2226.19,
                    "word": " Google",
                    "probability": 0.8623046875
                }
            ]
        },
        {
            "id": 387,
            "seek": 212375,
            "start": 2226.19,
            "end": 2229.31,
            "text": " to produce a lot of these statistical information",
            "tokens": [
                50364,
                281,
                5258,
                257,
                688,
                295,
                613,
                22820,
                1589,
                50464
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1923437485098839,
            "compression_ratio": 1.5258964143426295,
            "no_speech_prob": 0.0078125,
            "words": [
                {
                    "start": 2226.19,
                    "end": 2226.33,
                    "word": " to",
                    "probability": 0.409912109375
                },
                {
                    "start": 2226.33,
                    "end": 2226.75,
                    "word": " produce",
                    "probability": 0.6103515625
                },
                {
                    "start": 2226.97,
                    "end": 2227.09,
                    "word": " a lot",
                    "probability": 0.90185546875
                },
                {
                    "start": 2227.09,
                    "end": 2227.25,
                    "word": " of",
                    "probability": 0.89990234375
                },
                {
                    "start": 2227.25,
                    "end": 2227.57,
                    "word": " these",
                    "probability": 0.362548828125
                },
                {
                    "start": 2227.57,
                    "end": 2228.49,
                    "word": " statistical",
                    "probability": 0.8046875
                },
                {
                    "start": 2228.49,
                    "end": 2229.31,
                    "word": " information",
                    "probability": 0.73193359375
                }
            ]
        },
        {
            "id": 388,
            "seek": 212375,
            "start": 2229.31,
            "end": 2231.13,
            "text": " in a private manner.",
            "tokens": [
                50514,
                294,
                257,
                4551,
                9060,
                13,
                50564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1923437485098839,
            "compression_ratio": 1.5258964143426295,
            "no_speech_prob": 0.0078125,
            "words": [
                {
                    "start": 2229.31,
                    "end": 2229.71,
                    "word": " in",
                    "probability": 0.89892578125
                },
                {
                    "start": 2229.87,
                    "end": 2230.19,
                    "word": " a private",
                    "probability": 0.78515625
                },
                {
                    "start": 2230.19,
                    "end": 2230.55,
                    "word": " manner",
                    "probability": 0.7958984375
                },
                {
                    "start": 2230.55,
                    "end": 2231.13,
                    "word": ".",
                    "probability": 0.97021484375
                }
            ]
        },
        {
            "id": 389,
            "seek": 212375,
            "start": 2233.19,
            "end": 2237.45,
            "text": " Great. Any input from Aditi or Royce on that particular issue?",
            "tokens": [
                50714,
                3769,
                13,
                2639,
                4846,
                490,
                1999,
                8707,
                420,
                8751,
                384,
                322,
                300,
                1729,
                2734,
                30,
                50864
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1923437485098839,
            "compression_ratio": 1.5258964143426295,
            "no_speech_prob": 0.0078125,
            "words": [
                {
                    "start": 2233.19,
                    "end": 2233.77,
                    "word": " Great",
                    "probability": 0.41357421875
                },
                {
                    "start": 2233.77,
                    "end": 2233.87,
                    "word": ".",
                    "probability": 0.64306640625
                },
                {
                    "start": 2233.87,
                    "end": 2233.99,
                    "word": " Any",
                    "probability": 0.7900390625
                },
                {
                    "start": 2233.99,
                    "end": 2234.33,
                    "word": " input",
                    "probability": 0.87255859375
                },
                {
                    "start": 2234.33,
                    "end": 2234.69,
                    "word": " from",
                    "probability": 0.8203125
                },
                {
                    "start": 2234.69,
                    "end": 2234.99,
                    "word": " Aditi",
                    "probability": 0.713623046875
                },
                {
                    "start": 2234.99,
                    "end": 2235.31,
                    "word": " or",
                    "probability": 0.87353515625
                },
                {
                    "start": 2235.31,
                    "end": 2235.85,
                    "word": " Royce",
                    "probability": 0.910888671875
                },
                {
                    "start": 2235.85,
                    "end": 2235.99,
                    "word": " on",
                    "probability": 0.904296875
                },
                {
                    "start": 2235.99,
                    "end": 2236.19,
                    "word": " that",
                    "probability": 0.88134765625
                },
                {
                    "start": 2236.19,
                    "end": 2236.53,
                    "word": " particular",
                    "probability": 0.81005859375
                },
                {
                    "start": 2236.53,
                    "end": 2236.97,
                    "word": " issue",
                    "probability": 0.89599609375
                },
                {
                    "start": 2236.97,
                    "end": 2237.45,
                    "word": "?",
                    "probability": 0.98095703125
                }
            ]
        },
        {
            "id": 390,
            "seek": 212375,
            "start": 2237.45,
            "end": 2239.41,
            "text": " No, I think Miguel captured it.",
            "tokens": [
                50914,
                883,
                11,
                286,
                519,
                29150,
                11828,
                309,
                13,
                50964
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1923437485098839,
            "compression_ratio": 1.5258964143426295,
            "no_speech_prob": 0.0078125,
            "words": [
                {
                    "start": 2237.45,
                    "end": 2237.67,
                    "word": " No",
                    "probability": 0.8681640625
                },
                {
                    "start": 2237.67,
                    "end": 2237.73,
                    "word": ",",
                    "probability": 0.72998046875
                },
                {
                    "start": 2237.73,
                    "end": 2237.81,
                    "word": " I",
                    "probability": 0.9970703125
                },
                {
                    "start": 2237.81,
                    "end": 2237.95,
                    "word": " think",
                    "probability": 0.8203125
                },
                {
                    "start": 2237.95,
                    "end": 2238.17,
                    "word": " Miguel",
                    "probability": 0.80322265625
                },
                {
                    "start": 2238.17,
                    "end": 2238.55,
                    "word": " captured",
                    "probability": 0.65576171875
                },
                {
                    "start": 2238.55,
                    "end": 2238.91,
                    "word": " it",
                    "probability": 0.9033203125
                },
                {
                    "start": 2238.91,
                    "end": 2239.41,
                    "word": ".",
                    "probability": 0.9765625
                }
            ]
        },
        {
            "id": 391,
            "seek": 212375,
            "start": 2239.41,
            "end": 2243.53,
            "text": " Great. And so this is kind of a related question, I guess.",
            "tokens": [
                51014,
                3769,
                13,
                400,
                370,
                341,
                307,
                733,
                295,
                257,
                4077,
                1168,
                11,
                286,
                2041,
                13,
                51164
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1923437485098839,
            "compression_ratio": 1.5258964143426295,
            "no_speech_prob": 0.0078125,
            "words": [
                {
                    "start": 2239.41,
                    "end": 2239.69,
                    "word": " Great",
                    "probability": 0.8720703125
                },
                {
                    "start": 2239.69,
                    "end": 2240.27,
                    "word": ".",
                    "probability": 0.9404296875
                },
                {
                    "start": 2241.27,
                    "end": 2241.85,
                    "word": " And",
                    "probability": 0.75537109375
                },
                {
                    "start": 2241.85,
                    "end": 2242.05,
                    "word": " so",
                    "probability": 0.89501953125
                },
                {
                    "start": 2242.05,
                    "end": 2242.23,
                    "word": " this",
                    "probability": 0.7744140625
                },
                {
                    "start": 2242.23,
                    "end": 2242.35,
                    "word": " is",
                    "probability": 0.91015625
                },
                {
                    "start": 2242.35,
                    "end": 2242.49,
                    "word": " kind",
                    "probability": 0.7373046875
                },
                {
                    "start": 2242.49,
                    "end": 2242.57,
                    "word": " of",
                    "probability": 0.90087890625
                },
                {
                    "start": 2242.63,
                    "end": 2242.85,
                    "word": " a related",
                    "probability": 0.86328125
                },
                {
                    "start": 2242.85,
                    "end": 2243.21,
                    "word": " question",
                    "probability": 0.8037109375
                },
                {
                    "start": 2243.21,
                    "end": 2243.25,
                    "word": ",",
                    "probability": 0.80908203125
                },
                {
                    "start": 2243.25,
                    "end": 2243.35,
                    "word": " I",
                    "probability": 0.9921875
                },
                {
                    "start": 2243.35,
                    "end": 2243.47,
                    "word": " guess",
                    "probability": 0.90869140625
                },
                {
                    "start": 2243.47,
                    "end": 2243.53,
                    "word": ".",
                    "probability": 0.87646484375
                }
            ]
        },
        {
            "id": 392,
            "seek": 212375,
            "start": 2243.53,
            "end": 2246.01,
            "text": " Can you comment on what Google services,",
            "tokens": [
                51214,
                1664,
                291,
                2871,
                322,
                437,
                3329,
                3328,
                11,
                51314
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1923437485098839,
            "compression_ratio": 1.5258964143426295,
            "no_speech_prob": 0.0078125,
            "words": [
                {
                    "start": 2243.53,
                    "end": 2243.63,
                    "word": " Can",
                    "probability": 0.8984375
                },
                {
                    "start": 2243.63,
                    "end": 2243.77,
                    "word": " you",
                    "probability": 0.89404296875
                },
                {
                    "start": 2243.77,
                    "end": 2244.05,
                    "word": " comment",
                    "probability": 0.859375
                },
                {
                    "start": 2244.05,
                    "end": 2244.41,
                    "word": " on",
                    "probability": 0.9091796875
                },
                {
                    "start": 2244.41,
                    "end": 2244.83,
                    "word": " what",
                    "probability": 0.888671875
                },
                {
                    "start": 2244.83,
                    "end": 2245.09,
                    "word": " Google",
                    "probability": 0.86767578125
                },
                {
                    "start": 2245.09,
                    "end": 2245.57,
                    "word": " services",
                    "probability": 0.85009765625
                },
                {
                    "start": 2245.57,
                    "end": 2246.01,
                    "word": ",",
                    "probability": 0.6884765625
                }
            ]
        },
        {
            "id": 393,
            "seek": 212375,
            "start": 2246.01,
            "end": 2248.23,
            "text": " besides location visit and busy time reporting,",
            "tokens": [
                51364,
                11868,
                4914,
                3441,
                293,
                5856,
                565,
                10031,
                11,
                51414
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1923437485098839,
            "compression_ratio": 1.5258964143426295,
            "no_speech_prob": 0.0078125,
            "words": [
                {
                    "start": 2246.01,
                    "end": 2246.33,
                    "word": " besides",
                    "probability": 0.8486328125
                },
                {
                    "start": 2246.33,
                    "end": 2246.75,
                    "word": " location",
                    "probability": 0.55126953125
                },
                {
                    "start": 2246.75,
                    "end": 2247.05,
                    "word": " visit",
                    "probability": 0.7744140625
                },
                {
                    "start": 2247.05,
                    "end": 2247.19,
                    "word": " and",
                    "probability": 0.892578125
                },
                {
                    "start": 2247.19,
                    "end": 2247.37,
                    "word": " busy",
                    "probability": 0.80224609375
                },
                {
                    "start": 2247.37,
                    "end": 2247.61,
                    "word": " time",
                    "probability": 0.65234375
                },
                {
                    "start": 2247.61,
                    "end": 2248.01,
                    "word": " reporting",
                    "probability": 0.75830078125
                },
                {
                    "start": 2248.01,
                    "end": 2248.23,
                    "word": ",",
                    "probability": 0.94482421875
                }
            ]
        },
        {
            "id": 394,
            "seek": 212375,
            "start": 2248.23,
            "end": 2250.81,
            "text": " are implementing differential privacy today?",
            "tokens": [
                51464,
                366,
                18114,
                15756,
                11427,
                965,
                30,
                51564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1923437485098839,
            "compression_ratio": 1.5258964143426295,
            "no_speech_prob": 0.0078125,
            "words": [
                {
                    "start": 2248.23,
                    "end": 2248.79,
                    "word": " are",
                    "probability": 0.896484375
                },
                {
                    "start": 2248.79,
                    "end": 2249.25,
                    "word": " implementing",
                    "probability": 0.65771484375
                },
                {
                    "start": 2249.25,
                    "end": 2249.73,
                    "word": " differential",
                    "probability": 0.6787109375
                },
                {
                    "start": 2249.73,
                    "end": 2250.19,
                    "word": " privacy",
                    "probability": 0.84912109375
                },
                {
                    "start": 2250.19,
                    "end": 2250.65,
                    "word": " today",
                    "probability": 0.78369140625
                },
                {
                    "start": 2250.65,
                    "end": 2250.81,
                    "word": "?",
                    "probability": 0.97412109375
                }
            ]
        },
        {
            "id": 395,
            "seek": 212375,
            "start": 2250.81,
            "end": 2252.63,
            "text": " Can you give us a sense?",
            "tokens": [
                51614,
                1664,
                291,
                976,
                505,
                257,
                2020,
                30,
                51664
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1923437485098839,
            "compression_ratio": 1.5258964143426295,
            "no_speech_prob": 0.0078125,
            "words": [
                {
                    "start": 2250.81,
                    "end": 2250.99,
                    "word": " Can",
                    "probability": 0.88916015625
                },
                {
                    "start": 2250.99,
                    "end": 2251.07,
                    "word": " you",
                    "probability": 0.8876953125
                },
                {
                    "start": 2251.07,
                    "end": 2251.21,
                    "word": " give",
                    "probability": 0.8115234375
                },
                {
                    "start": 2251.21,
                    "end": 2251.37,
                    "word": " us",
                    "probability": 0.9052734375
                },
                {
                    "start": 2251.73,
                    "end": 2252.05,
                    "word": " a sense",
                    "probability": 0.720703125
                },
                {
                    "start": 2252.05,
                    "end": 2252.63,
                    "word": "?",
                    "probability": 0.99267578125
                }
            ]
        },
        {
            "id": 396,
            "seek": 215203,
            "start": 2253.57,
            "end": 2258.11,
            "text": " Yeah. So the mobility reports also use differential privacy",
            "tokens": [
                50414,
                865,
                13,
                407,
                264,
                16199,
                7122,
                611,
                764,
                15756,
                11427,
                50614
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20084269662921347,
            "compression_ratio": 1.775,
            "no_speech_prob": 0.004978179931640625,
            "words": [
                {
                    "start": 2253.57,
                    "end": 2253.93,
                    "word": " Yeah",
                    "probability": 0.1756591796875
                },
                {
                    "start": 2253.93,
                    "end": 2254.17,
                    "word": ".",
                    "probability": 0.1983642578125
                },
                {
                    "start": 2254.17,
                    "end": 2254.53,
                    "word": " So",
                    "probability": 0.87060546875
                },
                {
                    "start": 2254.53,
                    "end": 2255.33,
                    "word": " the",
                    "probability": 0.63330078125
                },
                {
                    "start": 2255.33,
                    "end": 2255.65,
                    "word": " mobility",
                    "probability": 0.4951171875
                },
                {
                    "start": 2255.65,
                    "end": 2256.29,
                    "word": " reports",
                    "probability": 0.7744140625
                },
                {
                    "start": 2256.29,
                    "end": 2257.03,
                    "word": " also",
                    "probability": 0.75927734375
                },
                {
                    "start": 2257.03,
                    "end": 2257.33,
                    "word": " use",
                    "probability": 0.791015625
                },
                {
                    "start": 2257.33,
                    "end": 2257.65,
                    "word": " differential",
                    "probability": 0.6357421875
                },
                {
                    "start": 2257.65,
                    "end": 2258.11,
                    "word": " privacy",
                    "probability": 0.7744140625
                }
            ]
        },
        {
            "id": 397,
            "seek": 215203,
            "start": 2258.11,
            "end": 2259.81,
            "text": " for all of their reporting.",
            "tokens": [
                50664,
                337,
                439,
                295,
                641,
                10031,
                13,
                50714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20084269662921347,
            "compression_ratio": 1.775,
            "no_speech_prob": 0.004978179931640625,
            "words": [
                {
                    "start": 2258.11,
                    "end": 2258.43,
                    "word": " for",
                    "probability": 0.89599609375
                },
                {
                    "start": 2258.43,
                    "end": 2258.65,
                    "word": " all",
                    "probability": 0.90185546875
                },
                {
                    "start": 2258.65,
                    "end": 2258.77,
                    "word": " of",
                    "probability": 0.90771484375
                },
                {
                    "start": 2258.77,
                    "end": 2258.91,
                    "word": " their",
                    "probability": 0.880859375
                },
                {
                    "start": 2258.91,
                    "end": 2259.45,
                    "word": " reporting",
                    "probability": 0.7646484375
                },
                {
                    "start": 2259.45,
                    "end": 2259.81,
                    "word": ".",
                    "probability": 0.97412109375
                }
            ]
        },
        {
            "id": 398,
            "seek": 215203,
            "start": 2260.15,
            "end": 2263.81,
            "text": " We have a lot of features on the backend",
            "tokens": [
                50764,
                492,
                362,
                257,
                688,
                295,
                4122,
                322,
                264,
                38087,
                50914
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20084269662921347,
            "compression_ratio": 1.775,
            "no_speech_prob": 0.004978179931640625,
            "words": [
                {
                    "start": 2260.15,
                    "end": 2260.59,
                    "word": " We",
                    "probability": 0.88671875
                },
                {
                    "start": 2260.59,
                    "end": 2261.23,
                    "word": " have",
                    "probability": 0.53076171875
                },
                {
                    "start": 2261.45,
                    "end": 2261.53,
                    "word": " a lot",
                    "probability": 0.89892578125
                },
                {
                    "start": 2261.53,
                    "end": 2261.67,
                    "word": " of",
                    "probability": 0.90576171875
                },
                {
                    "start": 2261.67,
                    "end": 2262.23,
                    "word": " features",
                    "probability": 0.7216796875
                },
                {
                    "start": 2262.23,
                    "end": 2263.29,
                    "word": " on",
                    "probability": 0.90087890625
                },
                {
                    "start": 2263.29,
                    "end": 2263.45,
                    "word": " the",
                    "probability": 0.81396484375
                },
                {
                    "start": 2263.45,
                    "end": 2263.81,
                    "word": " backend",
                    "probability": 0.69970703125
                }
            ]
        },
        {
            "id": 399,
            "seek": 215203,
            "start": 2263.81,
            "end": 2265.65,
            "text": " that use differential privacy.",
            "tokens": [
                50964,
                300,
                764,
                15756,
                11427,
                13,
                51014
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20084269662921347,
            "compression_ratio": 1.775,
            "no_speech_prob": 0.004978179931640625,
            "words": [
                {
                    "start": 2263.81,
                    "end": 2264.03,
                    "word": " that",
                    "probability": 0.86376953125
                },
                {
                    "start": 2264.03,
                    "end": 2264.25,
                    "word": " use",
                    "probability": 0.755859375
                },
                {
                    "start": 2264.25,
                    "end": 2264.65,
                    "word": " differential",
                    "probability": 0.75341796875
                },
                {
                    "start": 2264.65,
                    "end": 2265.15,
                    "word": " privacy",
                    "probability": 0.79052734375
                },
                {
                    "start": 2265.15,
                    "end": 2265.65,
                    "word": ".",
                    "probability": 0.98291015625
                }
            ]
        },
        {
            "id": 400,
            "seek": 215203,
            "start": 2265.65,
            "end": 2270.05,
            "text": " So for instance, Maps has this way of trying to cache tiles",
            "tokens": [
                51064,
                407,
                337,
                5197,
                11,
                28978,
                575,
                341,
                636,
                295,
                1382,
                281,
                19459,
                21982,
                51214
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20084269662921347,
            "compression_ratio": 1.775,
            "no_speech_prob": 0.004978179931640625,
            "words": [
                {
                    "start": 2265.65,
                    "end": 2265.77,
                    "word": " So",
                    "probability": 0.896484375
                },
                {
                    "start": 2265.77,
                    "end": 2265.99,
                    "word": " for",
                    "probability": 0.876953125
                },
                {
                    "start": 2265.99,
                    "end": 2266.27,
                    "word": " instance",
                    "probability": 0.9091796875
                },
                {
                    "start": 2266.27,
                    "end": 2266.43,
                    "word": ",",
                    "probability": 0.90625
                },
                {
                    "start": 2266.43,
                    "end": 2266.67,
                    "word": " Maps",
                    "probability": 0.5185546875
                },
                {
                    "start": 2266.67,
                    "end": 2267.41,
                    "word": " has",
                    "probability": 0.888671875
                },
                {
                    "start": 2267.41,
                    "end": 2267.63,
                    "word": " this",
                    "probability": 0.87158203125
                },
                {
                    "start": 2267.63,
                    "end": 2268.03,
                    "word": " way",
                    "probability": 0.90283203125
                },
                {
                    "start": 2268.03,
                    "end": 2268.47,
                    "word": " of",
                    "probability": 0.90380859375
                },
                {
                    "start": 2268.47,
                    "end": 2268.75,
                    "word": " trying",
                    "probability": 0.80517578125
                },
                {
                    "start": 2268.75,
                    "end": 2268.97,
                    "word": " to",
                    "probability": 0.90380859375
                },
                {
                    "start": 2268.97,
                    "end": 2269.47,
                    "word": " cache",
                    "probability": 0.7333984375
                },
                {
                    "start": 2269.47,
                    "end": 2270.05,
                    "word": " tiles",
                    "probability": 0.7568359375
                }
            ]
        },
        {
            "id": 401,
            "seek": 215203,
            "start": 2270.47,
            "end": 2274.63,
            "text": " that you'll probably use when you have no data.",
            "tokens": [
                51264,
                300,
                291,
                603,
                1391,
                764,
                562,
                291,
                362,
                572,
                1412,
                13,
                51414
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20084269662921347,
            "compression_ratio": 1.775,
            "no_speech_prob": 0.004978179931640625,
            "words": [
                {
                    "start": 2270.47,
                    "end": 2270.83,
                    "word": " that",
                    "probability": 0.86962890625
                },
                {
                    "start": 2270.83,
                    "end": 2271.19,
                    "word": " you'll",
                    "probability": 0.89501953125
                },
                {
                    "start": 2271.19,
                    "end": 2271.55,
                    "word": " probably",
                    "probability": 0.89794921875
                },
                {
                    "start": 2271.55,
                    "end": 2272.39,
                    "word": " use",
                    "probability": 0.8173828125
                },
                {
                    "start": 2272.39,
                    "end": 2272.99,
                    "word": " when",
                    "probability": 0.87060546875
                },
                {
                    "start": 2272.99,
                    "end": 2273.17,
                    "word": " you",
                    "probability": 0.8935546875
                },
                {
                    "start": 2273.17,
                    "end": 2273.43,
                    "word": " have",
                    "probability": 0.8759765625
                },
                {
                    "start": 2273.43,
                    "end": 2273.69,
                    "word": " no",
                    "probability": 0.8974609375
                },
                {
                    "start": 2273.69,
                    "end": 2274.05,
                    "word": " data",
                    "probability": 0.8271484375
                },
                {
                    "start": 2274.05,
                    "end": 2274.63,
                    "word": ".",
                    "probability": 0.7939453125
                }
            ]
        },
        {
            "id": 402,
            "seek": 215203,
            "start": 2274.63,
            "end": 2278.43,
            "text": " And the way that Maps does the ranking on the backend",
            "tokens": [
                51464,
                400,
                264,
                636,
                300,
                28978,
                775,
                264,
                17833,
                322,
                264,
                38087,
                51614
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20084269662921347,
            "compression_ratio": 1.775,
            "no_speech_prob": 0.004978179931640625,
            "words": [
                {
                    "start": 2274.63,
                    "end": 2274.81,
                    "word": " And",
                    "probability": 0.8681640625
                },
                {
                    "start": 2274.81,
                    "end": 2275.03,
                    "word": " the",
                    "probability": 0.8193359375
                },
                {
                    "start": 2275.03,
                    "end": 2275.29,
                    "word": " way",
                    "probability": 0.90185546875
                },
                {
                    "start": 2275.29,
                    "end": 2275.53,
                    "word": " that",
                    "probability": 0.88232421875
                },
                {
                    "start": 2275.53,
                    "end": 2276.27,
                    "word": " Maps",
                    "probability": 0.60546875
                },
                {
                    "start": 2276.27,
                    "end": 2276.63,
                    "word": " does",
                    "probability": 0.91259765625
                },
                {
                    "start": 2276.63,
                    "end": 2276.93,
                    "word": " the",
                    "probability": 0.8056640625
                },
                {
                    "start": 2276.93,
                    "end": 2277.31,
                    "word": " ranking",
                    "probability": 0.7890625
                },
                {
                    "start": 2277.31,
                    "end": 2277.89,
                    "word": " on",
                    "probability": 0.90625
                },
                {
                    "start": 2277.89,
                    "end": 2277.99,
                    "word": " the",
                    "probability": 0.81396484375
                },
                {
                    "start": 2277.99,
                    "end": 2278.43,
                    "word": " backend",
                    "probability": 0.892578125
                }
            ]
        },
        {
            "id": 403,
            "seek": 215203,
            "start": 2278.43,
            "end": 2280.91,
            "text": " is by using differential privacy.",
            "tokens": [
                51664,
                307,
                538,
                1228,
                15756,
                11427,
                13,
                51714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20084269662921347,
            "compression_ratio": 1.775,
            "no_speech_prob": 0.004978179931640625,
            "words": [
                {
                    "start": 2278.43,
                    "end": 2278.79,
                    "word": " is",
                    "probability": 0.87841796875
                },
                {
                    "start": 2278.79,
                    "end": 2278.99,
                    "word": " by",
                    "probability": 0.9111328125
                },
                {
                    "start": 2278.99,
                    "end": 2279.29,
                    "word": " using",
                    "probability": 0.830078125
                },
                {
                    "start": 2279.29,
                    "end": 2279.65,
                    "word": " differential",
                    "probability": 0.70703125
                },
                {
                    "start": 2279.65,
                    "end": 2280.23,
                    "word": " privacy",
                    "probability": 0.7216796875
                },
                {
                    "start": 2280.23,
                    "end": 2280.91,
                    "word": ".",
                    "probability": 0.92578125
                }
            ]
        },
        {
            "id": 404,
            "seek": 217707,
            "start": 2280.91,
            "end": 2283.85,
            "text": " Popular Dishes on Maps also uses differential privacy",
            "tokens": [
                50364,
                37637,
                413,
                16423,
                322,
                28978,
                611,
                4960,
                15756,
                11427,
                50464
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2943548348642165,
            "compression_ratio": 1.6563876651982379,
            "no_speech_prob": 0.0002199411392211914,
            "words": [
                {
                    "start": 2280.91,
                    "end": 2281.19,
                    "word": " Popular",
                    "probability": 0.830078125
                },
                {
                    "start": 2281.19,
                    "end": 2281.53,
                    "word": " Dishes",
                    "probability": 0.55047607421875
                },
                {
                    "start": 2281.53,
                    "end": 2281.83,
                    "word": " on",
                    "probability": 0.83056640625
                },
                {
                    "start": 2281.83,
                    "end": 2282.13,
                    "word": " Maps",
                    "probability": 0.81103515625
                },
                {
                    "start": 2282.13,
                    "end": 2282.45,
                    "word": " also",
                    "probability": 0.80859375
                },
                {
                    "start": 2282.45,
                    "end": 2282.87,
                    "word": " uses",
                    "probability": 0.81201171875
                },
                {
                    "start": 2282.87,
                    "end": 2283.37,
                    "word": " differential",
                    "probability": 0.5478515625
                },
                {
                    "start": 2283.37,
                    "end": 2283.85,
                    "word": " privacy",
                    "probability": 0.82177734375
                }
            ]
        },
        {
            "id": 405,
            "seek": 217707,
            "start": 2283.85,
            "end": 2286.23,
            "text": " to do the ranking of the dishes.",
            "tokens": [
                50514,
                281,
                360,
                264,
                17833,
                295,
                264,
                10814,
                13,
                50564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2943548348642165,
            "compression_ratio": 1.6563876651982379,
            "no_speech_prob": 0.0002199411392211914,
            "words": [
                {
                    "start": 2283.85,
                    "end": 2284.15,
                    "word": " to",
                    "probability": 0.90234375
                },
                {
                    "start": 2284.15,
                    "end": 2284.27,
                    "word": " do",
                    "probability": 0.49755859375
                },
                {
                    "start": 2284.27,
                    "end": 2284.41,
                    "word": " the",
                    "probability": 0.81494140625
                },
                {
                    "start": 2284.41,
                    "end": 2284.69,
                    "word": " ranking",
                    "probability": 0.8037109375
                },
                {
                    "start": 2284.69,
                    "end": 2285.13,
                    "word": " of",
                    "probability": 0.8818359375
                },
                {
                    "start": 2285.13,
                    "end": 2285.33,
                    "word": " the",
                    "probability": 0.79833984375
                },
                {
                    "start": 2285.33,
                    "end": 2285.63,
                    "word": " dishes",
                    "probability": 0.7578125
                },
                {
                    "start": 2285.63,
                    "end": 2286.23,
                    "word": ".",
                    "probability": 0.6669921875
                }
            ]
        },
        {
            "id": 406,
            "seek": 217707,
            "start": 2286.53,
            "end": 2289.57,
            "text": " And those are some of the user-facing features that we have.",
            "tokens": [
                50614,
                400,
                729,
                366,
                512,
                295,
                264,
                4195,
                12,
                44046,
                4122,
                300,
                321,
                362,
                13,
                50764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2943548348642165,
            "compression_ratio": 1.6563876651982379,
            "no_speech_prob": 0.0002199411392211914,
            "words": [
                {
                    "start": 2286.53,
                    "end": 2286.71,
                    "word": " And",
                    "probability": 0.81494140625
                },
                {
                    "start": 2286.71,
                    "end": 2286.93,
                    "word": " those",
                    "probability": 0.84619140625
                },
                {
                    "start": 2286.93,
                    "end": 2287.07,
                    "word": " are",
                    "probability": 0.89208984375
                },
                {
                    "start": 2287.07,
                    "end": 2287.25,
                    "word": " some",
                    "probability": 0.80322265625
                },
                {
                    "start": 2287.25,
                    "end": 2287.41,
                    "word": " of",
                    "probability": 0.89794921875
                },
                {
                    "start": 2287.41,
                    "end": 2287.65,
                    "word": " the",
                    "probability": 0.80908203125
                },
                {
                    "start": 2287.65,
                    "end": 2288.01,
                    "word": " user",
                    "probability": 0.80322265625
                },
                {
                    "start": 2288.01,
                    "end": 2288.31,
                    "word": "-facing",
                    "probability": 0.68505859375
                },
                {
                    "start": 2288.31,
                    "end": 2288.73,
                    "word": " features",
                    "probability": 0.71484375
                },
                {
                    "start": 2288.73,
                    "end": 2288.91,
                    "word": " that",
                    "probability": 0.8857421875
                },
                {
                    "start": 2288.91,
                    "end": 2289.07,
                    "word": " we",
                    "probability": 0.90966796875
                },
                {
                    "start": 2289.07,
                    "end": 2289.41,
                    "word": " have",
                    "probability": 0.88037109375
                },
                {
                    "start": 2289.41,
                    "end": 2289.57,
                    "word": ".",
                    "probability": 0.97314453125
                }
            ]
        },
        {
            "id": 407,
            "seek": 217707,
            "start": 2289.57,
            "end": 2294.77,
            "text": " However, we use differential privacy a lot for internal analysis.",
            "tokens": [
                50814,
                2908,
                11,
                321,
                764,
                15756,
                11427,
                257,
                688,
                337,
                6920,
                5215,
                13,
                51014
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2943548348642165,
            "compression_ratio": 1.6563876651982379,
            "no_speech_prob": 0.0002199411392211914,
            "words": [
                {
                    "start": 2289.57,
                    "end": 2289.91,
                    "word": " However",
                    "probability": 0.83544921875
                },
                {
                    "start": 2289.91,
                    "end": 2290.45,
                    "word": ",",
                    "probability": 0.97705078125
                },
                {
                    "start": 2290.45,
                    "end": 2291.13,
                    "word": " we",
                    "probability": 0.900390625
                },
                {
                    "start": 2291.13,
                    "end": 2291.47,
                    "word": " use",
                    "probability": 0.7958984375
                },
                {
                    "start": 2291.47,
                    "end": 2291.91,
                    "word": " differential",
                    "probability": 0.7265625
                },
                {
                    "start": 2291.91,
                    "end": 2292.31,
                    "word": " privacy",
                    "probability": 0.8095703125
                },
                {
                    "start": 2292.69,
                    "end": 2292.93,
                    "word": " a lot",
                    "probability": 0.89453125
                },
                {
                    "start": 2292.93,
                    "end": 2293.17,
                    "word": " for",
                    "probability": 0.8935546875
                },
                {
                    "start": 2293.17,
                    "end": 2293.55,
                    "word": " internal",
                    "probability": 0.72314453125
                },
                {
                    "start": 2293.55,
                    "end": 2294.17,
                    "word": " analysis",
                    "probability": 0.78955078125
                },
                {
                    "start": 2294.17,
                    "end": 2294.77,
                    "word": ".",
                    "probability": 0.22607421875
                }
            ]
        },
        {
            "id": 408,
            "seek": 217707,
            "start": 2294.77,
            "end": 2298.39,
            "text": " And that internal analysis informs new features.",
            "tokens": [
                51064,
                400,
                300,
                6920,
                5215,
                45320,
                777,
                4122,
                13,
                51214
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2943548348642165,
            "compression_ratio": 1.6563876651982379,
            "no_speech_prob": 0.0002199411392211914,
            "words": [
                {
                    "start": 2294.77,
                    "end": 2295.05,
                    "word": " And",
                    "probability": 0.8779296875
                },
                {
                    "start": 2295.05,
                    "end": 2295.25,
                    "word": " that",
                    "probability": 0.87939453125
                },
                {
                    "start": 2295.25,
                    "end": 2295.57,
                    "word": " internal",
                    "probability": 0.72509765625
                },
                {
                    "start": 2295.57,
                    "end": 2296.05,
                    "word": " analysis",
                    "probability": 0.7734375
                },
                {
                    "start": 2296.05,
                    "end": 2296.73,
                    "word": " informs",
                    "probability": 0.70849609375
                },
                {
                    "start": 2296.73,
                    "end": 2297.33,
                    "word": " new",
                    "probability": 0.7607421875
                },
                {
                    "start": 2297.33,
                    "end": 2297.79,
                    "word": " features",
                    "probability": 0.728515625
                },
                {
                    "start": 2297.79,
                    "end": 2298.39,
                    "word": ".",
                    "probability": 0.990234375
                }
            ]
        },
        {
            "id": 409,
            "seek": 217707,
            "start": 2301.19,
            "end": 2301.95,
            "text": " Great.",
            "tokens": [
                51264,
                3769,
                13,
                51314
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2943548348642165,
            "compression_ratio": 1.6563876651982379,
            "no_speech_prob": 0.0002199411392211914,
            "words": [
                {
                    "start": 2301.19,
                    "end": 2301.49,
                    "word": " Great",
                    "probability": 0.002994537353515625
                },
                {
                    "start": 2301.49,
                    "end": 2301.95,
                    "word": ".",
                    "probability": 0.68896484375
                }
            ]
        },
        {
            "id": 410,
            "seek": 217707,
            "start": 2301.95,
            "end": 2303.91,
            "text": " I'm going to use the link to the COVID mobility reports",
            "tokens": [
                51364,
                286,
                478,
                516,
                281,
                764,
                264,
                2113,
                281,
                264,
                4566,
                16199,
                7122,
                51414
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2943548348642165,
            "compression_ratio": 1.6563876651982379,
            "no_speech_prob": 0.0002199411392211914,
            "words": [
                {
                    "start": 2301.95,
                    "end": 2302.05,
                    "word": " I'm",
                    "probability": 0.5169677734375
                },
                {
                    "start": 2302.05,
                    "end": 2302.05,
                    "word": " going",
                    "probability": 0.459228515625
                },
                {
                    "start": 2302.05,
                    "end": 2302.05,
                    "word": " to",
                    "probability": 0.90087890625
                },
                {
                    "start": 2302.05,
                    "end": 2302.05,
                    "word": " use",
                    "probability": 0.28564453125
                },
                {
                    "start": 2302.05,
                    "end": 2302.29,
                    "word": " the",
                    "probability": 0.8115234375
                },
                {
                    "start": 2302.29,
                    "end": 2302.51,
                    "word": " link",
                    "probability": 0.8544921875
                },
                {
                    "start": 2302.51,
                    "end": 2302.63,
                    "word": " to",
                    "probability": 0.89794921875
                },
                {
                    "start": 2302.63,
                    "end": 2302.77,
                    "word": " the",
                    "probability": 0.80517578125
                },
                {
                    "start": 2302.77,
                    "end": 2302.99,
                    "word": " COVID",
                    "probability": 0.8359375
                },
                {
                    "start": 2302.99,
                    "end": 2303.39,
                    "word": " mobility",
                    "probability": 0.37939453125
                },
                {
                    "start": 2303.39,
                    "end": 2303.91,
                    "word": " reports",
                    "probability": 0.7724609375
                }
            ]
        },
        {
            "id": 411,
            "seek": 217707,
            "start": 2303.91,
            "end": 2307.41,
            "text": " in the Discord channel so people can reference it.",
            "tokens": [
                51464,
                294,
                264,
                32623,
                2269,
                370,
                561,
                393,
                6408,
                309,
                13,
                51564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.2943548348642165,
            "compression_ratio": 1.6563876651982379,
            "no_speech_prob": 0.0002199411392211914,
            "words": [
                {
                    "start": 2303.91,
                    "end": 2304.55,
                    "word": " in",
                    "probability": 0.888671875
                },
                {
                    "start": 2304.55,
                    "end": 2304.67,
                    "word": " the",
                    "probability": 0.80908203125
                },
                {
                    "start": 2304.67,
                    "end": 2304.89,
                    "word": " Discord",
                    "probability": 0.70263671875
                },
                {
                    "start": 2304.89,
                    "end": 2305.25,
                    "word": " channel",
                    "probability": 0.80126953125
                },
                {
                    "start": 2305.25,
                    "end": 2305.51,
                    "word": " so",
                    "probability": 0.580078125
                },
                {
                    "start": 2305.51,
                    "end": 2305.73,
                    "word": " people",
                    "probability": 0.92333984375
                },
                {
                    "start": 2305.73,
                    "end": 2305.91,
                    "word": " can",
                    "probability": 0.8759765625
                },
                {
                    "start": 2305.91,
                    "end": 2306.15,
                    "word": " reference",
                    "probability": 0.6328125
                },
                {
                    "start": 2306.15,
                    "end": 2306.81,
                    "word": " it",
                    "probability": 0.131103515625
                },
                {
                    "start": 2306.81,
                    "end": 2307.41,
                    "word": ".",
                    "probability": 0.9482421875
                }
            ]
        },
        {
            "id": 412,
            "seek": 220707,
            "start": 2309.01,
            "end": 2312.39,
            "text": " All the short links from your slide deck into Discord as well.",
            "tokens": [
                50414,
                1057,
                264,
                2099,
                6123,
                490,
                428,
                4137,
                9341,
                666,
                32623,
                382,
                731,
                13,
                50564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.23655224884631204,
            "compression_ratio": 1.6109090909090908,
            "no_speech_prob": 0.045135498046875,
            "words": [
                {
                    "start": 2309.01,
                    "end": 2309.45,
                    "word": " All",
                    "probability": 0.1676025390625
                },
                {
                    "start": 2309.45,
                    "end": 2309.61,
                    "word": " the",
                    "probability": 0.8115234375
                },
                {
                    "start": 2309.61,
                    "end": 2309.87,
                    "word": " short",
                    "probability": 0.9140625
                },
                {
                    "start": 2309.87,
                    "end": 2310.09,
                    "word": " links",
                    "probability": 0.71337890625
                },
                {
                    "start": 2310.09,
                    "end": 2310.37,
                    "word": " from",
                    "probability": 0.8203125
                },
                {
                    "start": 2310.37,
                    "end": 2310.57,
                    "word": " your",
                    "probability": 0.81396484375
                },
                {
                    "start": 2310.57,
                    "end": 2310.87,
                    "word": " slide",
                    "probability": 0.79443359375
                },
                {
                    "start": 2310.87,
                    "end": 2311.13,
                    "word": " deck",
                    "probability": 0.89892578125
                },
                {
                    "start": 2311.13,
                    "end": 2311.39,
                    "word": " into",
                    "probability": 0.0955810546875
                },
                {
                    "start": 2311.39,
                    "end": 2311.73,
                    "word": " Discord",
                    "probability": 0.4921875
                },
                {
                    "start": 2311.73,
                    "end": 2312.05,
                    "word": " as",
                    "probability": 0.8544921875
                },
                {
                    "start": 2312.05,
                    "end": 2312.25,
                    "word": " well",
                    "probability": 0.888671875
                },
                {
                    "start": 2312.25,
                    "end": 2312.39,
                    "word": ".",
                    "probability": 0.66015625
                }
            ]
        },
        {
            "id": 413,
            "seek": 220707,
            "start": 2312.39,
            "end": 2313.11,
            "text": " Fantastic.",
            "tokens": [
                50614,
                21320,
                13,
                50664
            ],
            "temperature": 0.0,
            "avg_logprob": -0.23655224884631204,
            "compression_ratio": 1.6109090909090908,
            "no_speech_prob": 0.045135498046875,
            "words": [
                {
                    "start": 2312.39,
                    "end": 2312.67,
                    "word": " Fantastic",
                    "probability": 0.81591796875
                },
                {
                    "start": 2312.67,
                    "end": 2313.11,
                    "word": ".",
                    "probability": 0.8828125
                }
            ]
        },
        {
            "id": 414,
            "seek": 220707,
            "start": 2315.11,
            "end": 2317.15,
            "text": " And then just more broadly,",
            "tokens": [
                50714,
                400,
                550,
                445,
                544,
                19511,
                11,
                50814
            ],
            "temperature": 0.0,
            "avg_logprob": -0.23655224884631204,
            "compression_ratio": 1.6109090909090908,
            "no_speech_prob": 0.045135498046875,
            "words": [
                {
                    "start": 2315.11,
                    "end": 2315.55,
                    "word": " And",
                    "probability": 0.7705078125
                },
                {
                    "start": 2315.55,
                    "end": 2315.81,
                    "word": " then",
                    "probability": 0.60400390625
                },
                {
                    "start": 2315.81,
                    "end": 2315.99,
                    "word": " just",
                    "probability": 0.802734375
                },
                {
                    "start": 2315.99,
                    "end": 2316.25,
                    "word": " more",
                    "probability": 0.88525390625
                },
                {
                    "start": 2316.25,
                    "end": 2317.01,
                    "word": " broadly",
                    "probability": 0.8544921875
                },
                {
                    "start": 2317.01,
                    "end": 2317.15,
                    "word": ",",
                    "probability": 0.75732421875
                }
            ]
        },
        {
            "id": 415,
            "seek": 220707,
            "start": 2317.15,
            "end": 2318.75,
            "text": " and this is sort of a big long-term question,",
            "tokens": [
                50864,
                293,
                341,
                307,
                1333,
                295,
                257,
                955,
                938,
                12,
                7039,
                1168,
                11,
                50914
            ],
            "temperature": 0.0,
            "avg_logprob": -0.23655224884631204,
            "compression_ratio": 1.6109090909090908,
            "no_speech_prob": 0.045135498046875,
            "words": [
                {
                    "start": 2317.15,
                    "end": 2317.21,
                    "word": " and",
                    "probability": 0.53857421875
                },
                {
                    "start": 2317.21,
                    "end": 2317.31,
                    "word": " this",
                    "probability": 0.87060546875
                },
                {
                    "start": 2317.31,
                    "end": 2317.39,
                    "word": " is",
                    "probability": 0.90576171875
                },
                {
                    "start": 2317.39,
                    "end": 2317.55,
                    "word": " sort",
                    "probability": 0.75830078125
                },
                {
                    "start": 2317.55,
                    "end": 2317.63,
                    "word": " of",
                    "probability": 0.9091796875
                },
                {
                    "start": 2317.79,
                    "end": 2317.99,
                    "word": " a big",
                    "probability": 0.88232421875
                },
                {
                    "start": 2317.99,
                    "end": 2318.23,
                    "word": " long",
                    "probability": 0.7431640625
                },
                {
                    "start": 2318.23,
                    "end": 2318.33,
                    "word": "-term",
                    "probability": 0.732666015625
                },
                {
                    "start": 2318.33,
                    "end": 2318.67,
                    "word": " question",
                    "probability": 0.79833984375
                },
                {
                    "start": 2318.67,
                    "end": 2318.75,
                    "word": ",",
                    "probability": 0.591796875
                }
            ]
        },
        {
            "id": 416,
            "seek": 220707,
            "start": 2318.75,
            "end": 2321.01,
            "text": " what do you think is next for differential privacy?",
            "tokens": [
                50964,
                437,
                360,
                291,
                519,
                307,
                958,
                337,
                15756,
                11427,
                30,
                51014
            ],
            "temperature": 0.0,
            "avg_logprob": -0.23655224884631204,
            "compression_ratio": 1.6109090909090908,
            "no_speech_prob": 0.045135498046875,
            "words": [
                {
                    "start": 2318.75,
                    "end": 2318.83,
                    "word": " what",
                    "probability": 0.8681640625
                },
                {
                    "start": 2318.83,
                    "end": 2318.93,
                    "word": " do",
                    "probability": 0.89990234375
                },
                {
                    "start": 2318.93,
                    "end": 2318.97,
                    "word": " you",
                    "probability": 0.89599609375
                },
                {
                    "start": 2318.97,
                    "end": 2319.11,
                    "word": " think",
                    "probability": 0.81689453125
                },
                {
                    "start": 2319.11,
                    "end": 2319.31,
                    "word": " is",
                    "probability": 0.90283203125
                },
                {
                    "start": 2319.31,
                    "end": 2319.65,
                    "word": " next",
                    "probability": 0.86083984375
                },
                {
                    "start": 2319.65,
                    "end": 2320.03,
                    "word": " for",
                    "probability": 0.90087890625
                },
                {
                    "start": 2320.03,
                    "end": 2320.47,
                    "word": " differential",
                    "probability": 0.642578125
                },
                {
                    "start": 2320.47,
                    "end": 2321.01,
                    "word": " privacy",
                    "probability": 0.83984375
                },
                {
                    "start": 2321.01,
                    "end": 2321.01,
                    "word": "?",
                    "probability": 0.7763671875
                }
            ]
        },
        {
            "id": 417,
            "seek": 220707,
            "start": 2321.01,
            "end": 2323.79,
            "text": " That could be sort of in terms of the academic work",
            "tokens": [
                51064,
                663,
                727,
                312,
                1333,
                295,
                294,
                2115,
                295,
                264,
                7778,
                589,
                51164
            ],
            "temperature": 0.0,
            "avg_logprob": -0.23655224884631204,
            "compression_ratio": 1.6109090909090908,
            "no_speech_prob": 0.045135498046875,
            "words": [
                {
                    "start": 2321.01,
                    "end": 2321.31,
                    "word": " That",
                    "probability": 0.81591796875
                },
                {
                    "start": 2321.31,
                    "end": 2321.49,
                    "word": " could",
                    "probability": 0.8173828125
                },
                {
                    "start": 2321.49,
                    "end": 2321.75,
                    "word": " be",
                    "probability": 0.90625
                },
                {
                    "start": 2321.75,
                    "end": 2321.97,
                    "word": " sort",
                    "probability": 0.7001953125
                },
                {
                    "start": 2321.97,
                    "end": 2322.09,
                    "word": " of",
                    "probability": 0.90380859375
                },
                {
                    "start": 2322.09,
                    "end": 2322.41,
                    "word": " in",
                    "probability": 0.8603515625
                },
                {
                    "start": 2322.41,
                    "end": 2322.77,
                    "word": " terms",
                    "probability": 0.77490234375
                },
                {
                    "start": 2322.77,
                    "end": 2322.87,
                    "word": " of",
                    "probability": 0.904296875
                },
                {
                    "start": 2322.87,
                    "end": 2322.99,
                    "word": " the",
                    "probability": 0.822265625
                },
                {
                    "start": 2322.99,
                    "end": 2323.31,
                    "word": " academic",
                    "probability": 0.86669921875
                },
                {
                    "start": 2323.31,
                    "end": 2323.79,
                    "word": " work",
                    "probability": 0.80712890625
                }
            ]
        },
        {
            "id": 418,
            "seek": 220707,
            "start": 2323.79,
            "end": 2327.51,
            "text": " or in terms of the applied work that you see your team going with",
            "tokens": [
                51214,
                420,
                294,
                2115,
                295,
                264,
                6456,
                589,
                300,
                291,
                536,
                428,
                1469,
                516,
                365,
                51314
            ],
            "temperature": 0.0,
            "avg_logprob": -0.23655224884631204,
            "compression_ratio": 1.6109090909090908,
            "no_speech_prob": 0.045135498046875,
            "words": [
                {
                    "start": 2323.79,
                    "end": 2323.97,
                    "word": " or",
                    "probability": 0.7177734375
                },
                {
                    "start": 2323.97,
                    "end": 2324.17,
                    "word": " in",
                    "probability": 0.90869140625
                },
                {
                    "start": 2324.17,
                    "end": 2324.41,
                    "word": " terms",
                    "probability": 0.76953125
                },
                {
                    "start": 2324.41,
                    "end": 2324.65,
                    "word": " of",
                    "probability": 0.90673828125
                },
                {
                    "start": 2324.65,
                    "end": 2324.91,
                    "word": " the",
                    "probability": 0.81591796875
                },
                {
                    "start": 2324.91,
                    "end": 2325.29,
                    "word": " applied",
                    "probability": 0.67333984375
                },
                {
                    "start": 2325.29,
                    "end": 2325.53,
                    "word": " work",
                    "probability": 0.80810546875
                },
                {
                    "start": 2325.53,
                    "end": 2325.67,
                    "word": " that",
                    "probability": 0.88330078125
                },
                {
                    "start": 2325.67,
                    "end": 2325.81,
                    "word": " you",
                    "probability": 0.8994140625
                },
                {
                    "start": 2325.81,
                    "end": 2326.21,
                    "word": " see",
                    "probability": 0.82080078125
                },
                {
                    "start": 2326.21,
                    "end": 2326.75,
                    "word": " your",
                    "probability": 0.79833984375
                },
                {
                    "start": 2326.75,
                    "end": 2326.99,
                    "word": " team",
                    "probability": 0.91796875
                },
                {
                    "start": 2326.99,
                    "end": 2327.27,
                    "word": " going",
                    "probability": 0.84814453125
                },
                {
                    "start": 2327.27,
                    "end": 2327.51,
                    "word": " with",
                    "probability": 0.78369140625
                }
            ]
        },
        {
            "id": 419,
            "seek": 220707,
            "start": 2327.51,
            "end": 2328.91,
            "text": " in the next several years?",
            "tokens": [
                51364,
                294,
                264,
                958,
                2940,
                924,
                30,
                51414
            ],
            "temperature": 0.0,
            "avg_logprob": -0.23655224884631204,
            "compression_ratio": 1.6109090909090908,
            "no_speech_prob": 0.045135498046875,
            "words": [
                {
                    "start": 2327.51,
                    "end": 2327.63,
                    "word": " in",
                    "probability": 0.89404296875
                },
                {
                    "start": 2327.63,
                    "end": 2327.73,
                    "word": " the",
                    "probability": 0.81640625
                },
                {
                    "start": 2327.73,
                    "end": 2327.89,
                    "word": " next",
                    "probability": 0.8515625
                },
                {
                    "start": 2327.89,
                    "end": 2328.11,
                    "word": " several",
                    "probability": 0.8447265625
                },
                {
                    "start": 2328.11,
                    "end": 2328.47,
                    "word": " years",
                    "probability": 0.87939453125
                },
                {
                    "start": 2328.47,
                    "end": 2328.91,
                    "word": "?",
                    "probability": 0.477783203125
                }
            ]
        },
        {
            "id": 420,
            "seek": 220707,
            "start": 2329.47,
            "end": 2331.11,
            "text": " Royce, do you want to take that question?",
            "tokens": [
                51464,
                8751,
                384,
                11,
                360,
                291,
                528,
                281,
                747,
                300,
                1168,
                30,
                51514
            ],
            "temperature": 0.0,
            "avg_logprob": -0.23655224884631204,
            "compression_ratio": 1.6109090909090908,
            "no_speech_prob": 0.045135498046875,
            "words": [
                {
                    "start": 2329.47,
                    "end": 2329.73,
                    "word": " Royce",
                    "probability": 0.60931396484375
                },
                {
                    "start": 2329.73,
                    "end": 2329.73,
                    "word": ",",
                    "probability": 0.912109375
                },
                {
                    "start": 2329.73,
                    "end": 2329.85,
                    "word": " do",
                    "probability": 0.8935546875
                },
                {
                    "start": 2329.85,
                    "end": 2329.89,
                    "word": " you",
                    "probability": 0.89404296875
                },
                {
                    "start": 2329.89,
                    "end": 2329.99,
                    "word": " want",
                    "probability": 0.6982421875
                },
                {
                    "start": 2329.99,
                    "end": 2330.09,
                    "word": " to",
                    "probability": 0.904296875
                },
                {
                    "start": 2330.09,
                    "end": 2330.17,
                    "word": " take",
                    "probability": 0.80908203125
                },
                {
                    "start": 2330.17,
                    "end": 2330.33,
                    "word": " that",
                    "probability": 0.87255859375
                },
                {
                    "start": 2330.33,
                    "end": 2330.67,
                    "word": " question",
                    "probability": 0.79833984375
                },
                {
                    "start": 2330.67,
                    "end": 2331.11,
                    "word": "?",
                    "probability": 0.99267578125
                }
            ]
        },
        {
            "id": 421,
            "seek": 220707,
            "start": 2331.61,
            "end": 2335.15,
            "text": " Yeah.",
            "tokens": [
                51564,
                865,
                13,
                51614
            ],
            "temperature": 0.0,
            "avg_logprob": -0.23655224884631204,
            "compression_ratio": 1.6109090909090908,
            "no_speech_prob": 0.045135498046875,
            "words": [
                {
                    "start": 2331.61,
                    "end": 2331.61,
                    "word": " Yeah",
                    "probability": 0.2164306640625
                },
                {
                    "start": 2334.71,
                    "end": 2335.15,
                    "word": ".",
                    "probability": 0.9248046875
                }
            ]
        },
        {
            "id": 422,
            "seek": 220707,
            "start": 2337.27,
            "end": 2339.91,
            "text": " Miguel might be more suited for this one, I think.",
            "tokens": [
                51664,
                29150,
                1062,
                312,
                544,
                24736,
                337,
                341,
                472,
                11,
                286,
                519,
                13,
                51814
            ],
            "temperature": 0.0,
            "avg_logprob": -0.23655224884631204,
            "compression_ratio": 1.6109090909090908,
            "no_speech_prob": 0.045135498046875,
            "words": [
                {
                    "start": 2337.27,
                    "end": 2337.71,
                    "word": " Miguel",
                    "probability": 0.69873046875
                },
                {
                    "start": 2337.71,
                    "end": 2337.93,
                    "word": " might",
                    "probability": 0.80419921875
                },
                {
                    "start": 2337.93,
                    "end": 2338.09,
                    "word": " be",
                    "probability": 0.9013671875
                },
                {
                    "start": 2338.09,
                    "end": 2338.29,
                    "word": " more",
                    "probability": 0.88427734375
                },
                {
                    "start": 2338.29,
                    "end": 2338.53,
                    "word": " suited",
                    "probability": 0.90087890625
                },
                {
                    "start": 2338.53,
                    "end": 2338.73,
                    "word": " for",
                    "probability": 0.89990234375
                },
                {
                    "start": 2338.73,
                    "end": 2338.93,
                    "word": " this",
                    "probability": 0.880859375
                },
                {
                    "start": 2338.93,
                    "end": 2339.15,
                    "word": " one",
                    "probability": 0.8203125
                },
                {
                    "start": 2339.15,
                    "end": 2339.25,
                    "word": ",",
                    "probability": 0.80517578125
                },
                {
                    "start": 2339.25,
                    "end": 2339.47,
                    "word": " I",
                    "probability": 0.9921875
                },
                {
                    "start": 2339.47,
                    "end": 2339.75,
                    "word": " think",
                    "probability": 0.8125
                },
                {
                    "start": 2339.75,
                    "end": 2339.91,
                    "word": ".",
                    "probability": 0.62353515625
                }
            ]
        },
        {
            "id": 423,
            "seek": 223707,
            "start": 2342.23,
            "end": 2343.65,
            "text": " Do you want to...?",
            "tokens": [
                50414,
                1144,
                291,
                528,
                281,
                8964,
                50464
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20026882104976204,
            "compression_ratio": 1.6428571428571428,
            "no_speech_prob": 0.0024318695068359375,
            "words": [
                {
                    "start": 2342.23,
                    "end": 2342.77,
                    "word": " Do",
                    "probability": 0.2017822265625
                },
                {
                    "start": 2342.77,
                    "end": 2342.81,
                    "word": " you",
                    "probability": 0.896484375
                },
                {
                    "start": 2342.81,
                    "end": 2342.93,
                    "word": " want",
                    "probability": 0.389892578125
                },
                {
                    "start": 2342.93,
                    "end": 2343.65,
                    "word": " to...?",
                    "probability": 0.470489501953125
                }
            ]
        },
        {
            "id": 424,
            "seek": 223707,
            "start": 2343.65,
            "end": 2344.43,
            "text": " Okay.",
            "tokens": [
                50514,
                1033,
                13,
                50564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20026882104976204,
            "compression_ratio": 1.6428571428571428,
            "no_speech_prob": 0.0024318695068359375,
            "words": [
                {
                    "start": 2343.65,
                    "end": 2343.89,
                    "word": " Okay",
                    "probability": 0.328857421875
                },
                {
                    "start": 2343.89,
                    "end": 2344.43,
                    "word": ".",
                    "probability": 0.342041015625
                }
            ]
        },
        {
            "id": 425,
            "seek": 223707,
            "start": 2344.57,
            "end": 2349.49,
            "text": " I think that the biggest academic frontier",
            "tokens": [
                50614,
                286,
                519,
                300,
                264,
                3880,
                7778,
                35853,
                50764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20026882104976204,
            "compression_ratio": 1.6428571428571428,
            "no_speech_prob": 0.0024318695068359375,
            "words": [
                {
                    "start": 2344.57,
                    "end": 2344.81,
                    "word": " I",
                    "probability": 0.9580078125
                },
                {
                    "start": 2344.81,
                    "end": 2345.11,
                    "word": " think",
                    "probability": 0.80908203125
                },
                {
                    "start": 2345.11,
                    "end": 2347.53,
                    "word": " that",
                    "probability": 0.56298828125
                },
                {
                    "start": 2347.53,
                    "end": 2347.99,
                    "word": " the",
                    "probability": 0.8056640625
                },
                {
                    "start": 2347.99,
                    "end": 2348.31,
                    "word": " biggest",
                    "probability": 0.8935546875
                },
                {
                    "start": 2348.31,
                    "end": 2348.93,
                    "word": " academic",
                    "probability": 0.59033203125
                },
                {
                    "start": 2348.93,
                    "end": 2349.49,
                    "word": " frontier",
                    "probability": 0.82568359375
                }
            ]
        },
        {
            "id": 426,
            "seek": 223707,
            "start": 2349.49,
            "end": 2355.99,
            "text": " is improving the performance of differential privacy.",
            "tokens": [
                50814,
                307,
                11470,
                264,
                3389,
                295,
                15756,
                11427,
                13,
                51064
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20026882104976204,
            "compression_ratio": 1.6428571428571428,
            "no_speech_prob": 0.0024318695068359375,
            "words": [
                {
                    "start": 2349.49,
                    "end": 2349.93,
                    "word": " is",
                    "probability": 0.88818359375
                },
                {
                    "start": 2349.93,
                    "end": 2351.87,
                    "word": " improving",
                    "probability": 0.444580078125
                },
                {
                    "start": 2351.87,
                    "end": 2352.11,
                    "word": " the",
                    "probability": 0.80517578125
                },
                {
                    "start": 2352.11,
                    "end": 2354.49,
                    "word": " performance",
                    "probability": 0.7001953125
                },
                {
                    "start": 2354.49,
                    "end": 2354.71,
                    "word": " of",
                    "probability": 0.89599609375
                },
                {
                    "start": 2354.71,
                    "end": 2355.03,
                    "word": " differential",
                    "probability": 0.64111328125
                },
                {
                    "start": 2355.03,
                    "end": 2355.51,
                    "word": " privacy",
                    "probability": 0.83447265625
                },
                {
                    "start": 2355.51,
                    "end": 2355.99,
                    "word": ".",
                    "probability": 0.931640625
                }
            ]
        },
        {
            "id": 427,
            "seek": 223707,
            "start": 2355.99,
            "end": 2357.73,
            "text": " At the moment, for certain use cases,",
            "tokens": [
                51114,
                1711,
                264,
                1623,
                11,
                337,
                1629,
                764,
                3331,
                11,
                51164
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20026882104976204,
            "compression_ratio": 1.6428571428571428,
            "no_speech_prob": 0.0024318695068359375,
            "words": [
                {
                    "start": 2355.99,
                    "end": 2356.17,
                    "word": " At",
                    "probability": 0.90869140625
                },
                {
                    "start": 2356.17,
                    "end": 2356.29,
                    "word": " the",
                    "probability": 0.81591796875
                },
                {
                    "start": 2356.29,
                    "end": 2356.53,
                    "word": " moment",
                    "probability": 0.92529296875
                },
                {
                    "start": 2356.53,
                    "end": 2356.63,
                    "word": ",",
                    "probability": 0.59423828125
                },
                {
                    "start": 2356.63,
                    "end": 2356.75,
                    "word": " for",
                    "probability": 0.8837890625
                },
                {
                    "start": 2356.75,
                    "end": 2357.03,
                    "word": " certain",
                    "probability": 0.80712890625
                },
                {
                    "start": 2357.03,
                    "end": 2357.29,
                    "word": " use",
                    "probability": 0.78076171875
                },
                {
                    "start": 2357.29,
                    "end": 2357.53,
                    "word": " cases",
                    "probability": 0.78564453125
                },
                {
                    "start": 2357.53,
                    "end": 2357.73,
                    "word": ",",
                    "probability": 0.75732421875
                }
            ]
        },
        {
            "id": 428,
            "seek": 223707,
            "start": 2357.73,
            "end": 2360.63,
            "text": " you get very big hits in terms of utility.",
            "tokens": [
                51214,
                291,
                483,
                588,
                955,
                8664,
                294,
                2115,
                295,
                14877,
                13,
                51314
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20026882104976204,
            "compression_ratio": 1.6428571428571428,
            "no_speech_prob": 0.0024318695068359375,
            "words": [
                {
                    "start": 2357.73,
                    "end": 2357.77,
                    "word": " you",
                    "probability": 0.87646484375
                },
                {
                    "start": 2357.77,
                    "end": 2358.05,
                    "word": " get",
                    "probability": 0.88330078125
                },
                {
                    "start": 2358.05,
                    "end": 2358.71,
                    "word": " very",
                    "probability": 0.720703125
                },
                {
                    "start": 2358.71,
                    "end": 2358.99,
                    "word": " big",
                    "probability": 0.875
                },
                {
                    "start": 2358.99,
                    "end": 2359.31,
                    "word": " hits",
                    "probability": 0.87158203125
                },
                {
                    "start": 2359.31,
                    "end": 2359.49,
                    "word": " in",
                    "probability": 0.87451171875
                },
                {
                    "start": 2359.49,
                    "end": 2359.71,
                    "word": " terms",
                    "probability": 0.77197265625
                },
                {
                    "start": 2359.71,
                    "end": 2359.85,
                    "word": " of",
                    "probability": 0.90625
                },
                {
                    "start": 2359.85,
                    "end": 2360.25,
                    "word": " utility",
                    "probability": 0.75830078125
                },
                {
                    "start": 2360.25,
                    "end": 2360.63,
                    "word": ".",
                    "probability": 0.11090087890625
                }
            ]
        },
        {
            "id": 429,
            "seek": 223707,
            "start": 2360.63,
            "end": 2363.77,
            "text": " And therefore, it's not good for certain use cases.",
            "tokens": [
                51364,
                400,
                4412,
                11,
                309,
                311,
                406,
                665,
                337,
                1629,
                764,
                3331,
                13,
                51464
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20026882104976204,
            "compression_ratio": 1.6428571428571428,
            "no_speech_prob": 0.0024318695068359375,
            "words": [
                {
                    "start": 2360.63,
                    "end": 2360.81,
                    "word": " And",
                    "probability": 0.8427734375
                },
                {
                    "start": 2360.81,
                    "end": 2361.25,
                    "word": " therefore",
                    "probability": 0.673828125
                },
                {
                    "start": 2361.25,
                    "end": 2361.73,
                    "word": ",",
                    "probability": 0.845703125
                },
                {
                    "start": 2361.73,
                    "end": 2362.29,
                    "word": " it's",
                    "probability": 0.947265625
                },
                {
                    "start": 2362.29,
                    "end": 2362.45,
                    "word": " not",
                    "probability": 0.89013671875
                },
                {
                    "start": 2362.45,
                    "end": 2362.63,
                    "word": " good",
                    "probability": 0.89306640625
                },
                {
                    "start": 2362.63,
                    "end": 2362.81,
                    "word": " for",
                    "probability": 0.90087890625
                },
                {
                    "start": 2362.81,
                    "end": 2363.11,
                    "word": " certain",
                    "probability": 0.8076171875
                },
                {
                    "start": 2363.11,
                    "end": 2363.33,
                    "word": " use",
                    "probability": 0.79833984375
                },
                {
                    "start": 2363.33,
                    "end": 2363.61,
                    "word": " cases",
                    "probability": 0.80712890625
                },
                {
                    "start": 2363.61,
                    "end": 2363.77,
                    "word": ".",
                    "probability": 0.96826171875
                }
            ]
        },
        {
            "id": 430,
            "seek": 223707,
            "start": 2363.77,
            "end": 2367.01,
            "text": " So I really hope that within the academic community,",
            "tokens": [
                51514,
                407,
                286,
                534,
                1454,
                300,
                1951,
                264,
                7778,
                1768,
                11,
                51614
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20026882104976204,
            "compression_ratio": 1.6428571428571428,
            "no_speech_prob": 0.0024318695068359375,
            "words": [
                {
                    "start": 2363.77,
                    "end": 2363.87,
                    "word": " So",
                    "probability": 0.884765625
                },
                {
                    "start": 2363.87,
                    "end": 2364.05,
                    "word": " I",
                    "probability": 0.8076171875
                },
                {
                    "start": 2364.05,
                    "end": 2364.27,
                    "word": " really",
                    "probability": 0.78125
                },
                {
                    "start": 2364.27,
                    "end": 2364.59,
                    "word": " hope",
                    "probability": 0.8271484375
                },
                {
                    "start": 2364.59,
                    "end": 2364.87,
                    "word": " that",
                    "probability": 0.88330078125
                },
                {
                    "start": 2364.87,
                    "end": 2365.89,
                    "word": " within",
                    "probability": 0.6982421875
                },
                {
                    "start": 2365.89,
                    "end": 2366.09,
                    "word": " the",
                    "probability": 0.8134765625
                },
                {
                    "start": 2366.09,
                    "end": 2366.31,
                    "word": " academic",
                    "probability": 0.5673828125
                },
                {
                    "start": 2366.31,
                    "end": 2366.79,
                    "word": " community",
                    "probability": 0.8203125
                },
                {
                    "start": 2366.79,
                    "end": 2367.01,
                    "word": ",",
                    "probability": 0.80126953125
                }
            ]
        },
        {
            "id": 431,
            "seek": 223707,
            "start": 2367.01,
            "end": 2370.37,
            "text": " they can come up with new ways of doing differential privacy",
            "tokens": [
                51664,
                436,
                393,
                808,
                493,
                365,
                777,
                2098,
                295,
                884,
                15756,
                11427,
                51814
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20026882104976204,
            "compression_ratio": 1.6428571428571428,
            "no_speech_prob": 0.0024318695068359375,
            "words": [
                {
                    "start": 2367.01,
                    "end": 2367.57,
                    "word": " they",
                    "probability": 0.6845703125
                },
                {
                    "start": 2367.57,
                    "end": 2367.77,
                    "word": " can",
                    "probability": 0.89794921875
                },
                {
                    "start": 2367.77,
                    "end": 2367.97,
                    "word": " come",
                    "probability": 0.8134765625
                },
                {
                    "start": 2367.97,
                    "end": 2368.17,
                    "word": " up",
                    "probability": 0.9013671875
                },
                {
                    "start": 2368.17,
                    "end": 2368.43,
                    "word": " with",
                    "probability": 0.8125
                },
                {
                    "start": 2368.43,
                    "end": 2368.75,
                    "word": " new",
                    "probability": 0.81103515625
                },
                {
                    "start": 2368.75,
                    "end": 2369.09,
                    "word": " ways",
                    "probability": 0.80908203125
                },
                {
                    "start": 2369.09,
                    "end": 2369.27,
                    "word": " of",
                    "probability": 0.90625
                },
                {
                    "start": 2369.27,
                    "end": 2369.49,
                    "word": " doing",
                    "probability": 0.91552734375
                },
                {
                    "start": 2369.49,
                    "end": 2369.85,
                    "word": " differential",
                    "probability": 0.697265625
                },
                {
                    "start": 2369.85,
                    "end": 2370.37,
                    "word": " privacy",
                    "probability": 0.83203125
                }
            ]
        },
        {
            "id": 432,
            "seek": 226471,
            "start": 2370.51,
            "end": 2373.39,
            "text": " with less utility trade-offs.",
            "tokens": [
                50364,
                365,
                1570,
                14877,
                4923,
                12,
                19231,
                13,
                50464
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1659962800469729,
            "compression_ratio": 1.6234309623430963,
            "no_speech_prob": 0.0004868507385253906,
            "words": [
                {
                    "start": 2370.51,
                    "end": 2371.03,
                    "word": " with",
                    "probability": 0.611328125
                },
                {
                    "start": 2371.03,
                    "end": 2371.41,
                    "word": " less",
                    "probability": 0.88720703125
                },
                {
                    "start": 2371.41,
                    "end": 2372.01,
                    "word": " utility",
                    "probability": 0.81591796875
                },
                {
                    "start": 2372.01,
                    "end": 2372.53,
                    "word": " trade",
                    "probability": 0.8349609375
                },
                {
                    "start": 2372.53,
                    "end": 2372.79,
                    "word": "-offs",
                    "probability": 0.771728515625
                },
                {
                    "start": 2372.79,
                    "end": 2373.39,
                    "word": ".",
                    "probability": 0.83642578125
                }
            ]
        },
        {
            "id": 433,
            "seek": 226471,
            "start": 2373.39,
            "end": 2376.19,
            "text": " I think that's the other thing that I was thinking about",
            "tokens": [
                50514,
                286,
                519,
                300,
                311,
                264,
                661,
                551,
                300,
                286,
                390,
                1953,
                466,
                50614
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1659962800469729,
            "compression_ratio": 1.6234309623430963,
            "no_speech_prob": 0.0004868507385253906,
            "words": [
                {
                    "start": 2373.39,
                    "end": 2373.73,
                    "word": " I",
                    "probability": 0.9228515625
                },
                {
                    "start": 2373.73,
                    "end": 2373.95,
                    "word": " think",
                    "probability": 0.80517578125
                },
                {
                    "start": 2373.95,
                    "end": 2374.29,
                    "word": " that's",
                    "probability": 0.83984375
                },
                {
                    "start": 2374.29,
                    "end": 2374.83,
                    "word": " the",
                    "probability": 0.7509765625
                },
                {
                    "start": 2374.83,
                    "end": 2374.99,
                    "word": " other",
                    "probability": 0.76513671875
                },
                {
                    "start": 2374.99,
                    "end": 2375.25,
                    "word": " thing",
                    "probability": 0.81787109375
                },
                {
                    "start": 2375.25,
                    "end": 2375.37,
                    "word": " that",
                    "probability": 0.826171875
                },
                {
                    "start": 2375.37,
                    "end": 2375.47,
                    "word": " I",
                    "probability": 0.9951171875
                },
                {
                    "start": 2375.47,
                    "end": 2375.61,
                    "word": " was",
                    "probability": 0.8994140625
                },
                {
                    "start": 2375.61,
                    "end": 2375.83,
                    "word": " thinking",
                    "probability": 0.7451171875
                },
                {
                    "start": 2375.83,
                    "end": 2376.19,
                    "word": " about",
                    "probability": 0.818359375
                }
            ]
        },
        {
            "id": 434,
            "seek": 226471,
            "start": 2376.19,
            "end": 2380.57,
            "text": " is just make it generally more usable.",
            "tokens": [
                50664,
                307,
                445,
                652,
                309,
                5101,
                544,
                29975,
                13,
                50814
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1659962800469729,
            "compression_ratio": 1.6234309623430963,
            "no_speech_prob": 0.0004868507385253906,
            "words": [
                {
                    "start": 2376.19,
                    "end": 2376.75,
                    "word": " is",
                    "probability": 0.634765625
                },
                {
                    "start": 2376.75,
                    "end": 2377.49,
                    "word": " just",
                    "probability": 0.75390625
                },
                {
                    "start": 2377.49,
                    "end": 2377.87,
                    "word": " make",
                    "probability": 0.79833984375
                },
                {
                    "start": 2377.87,
                    "end": 2378.13,
                    "word": " it",
                    "probability": 0.9013671875
                },
                {
                    "start": 2378.13,
                    "end": 2379.27,
                    "word": " generally",
                    "probability": 0.837890625
                },
                {
                    "start": 2379.27,
                    "end": 2379.63,
                    "word": " more",
                    "probability": 0.87890625
                },
                {
                    "start": 2379.63,
                    "end": 2380.21,
                    "word": " usable",
                    "probability": 0.81884765625
                },
                {
                    "start": 2380.21,
                    "end": 2380.57,
                    "word": ".",
                    "probability": 0.7841796875
                }
            ]
        },
        {
            "id": 435,
            "seek": 226471,
            "start": 2380.57,
            "end": 2384.13,
            "text": " At the moment, it requires a significant level of expertise",
            "tokens": [
                50864,
                1711,
                264,
                1623,
                11,
                309,
                7029,
                257,
                4776,
                1496,
                295,
                11769,
                51014
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1659962800469729,
            "compression_ratio": 1.6234309623430963,
            "no_speech_prob": 0.0004868507385253906,
            "words": [
                {
                    "start": 2380.57,
                    "end": 2381.01,
                    "word": " At",
                    "probability": 0.90478515625
                },
                {
                    "start": 2381.01,
                    "end": 2381.15,
                    "word": " the",
                    "probability": 0.81640625
                },
                {
                    "start": 2381.15,
                    "end": 2381.43,
                    "word": " moment",
                    "probability": 0.92724609375
                },
                {
                    "start": 2381.43,
                    "end": 2381.77,
                    "word": ",",
                    "probability": 0.66552734375
                },
                {
                    "start": 2381.77,
                    "end": 2382.27,
                    "word": " it",
                    "probability": 0.89599609375
                },
                {
                    "start": 2382.27,
                    "end": 2382.59,
                    "word": " requires",
                    "probability": 0.75146484375
                },
                {
                    "start": 2382.75,
                    "end": 2383.05,
                    "word": " a significant",
                    "probability": 0.8173828125
                },
                {
                    "start": 2383.05,
                    "end": 2383.35,
                    "word": " level",
                    "probability": 0.80712890625
                },
                {
                    "start": 2383.35,
                    "end": 2383.57,
                    "word": " of",
                    "probability": 0.8984375
                },
                {
                    "start": 2383.57,
                    "end": 2384.13,
                    "word": " expertise",
                    "probability": 0.83251953125
                }
            ]
        },
        {
            "id": 436,
            "seek": 226471,
            "start": 2384.13,
            "end": 2386.85,
            "text": " to understand it and to implement it.",
            "tokens": [
                51064,
                281,
                1223,
                309,
                293,
                281,
                4445,
                309,
                13,
                51114
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1659962800469729,
            "compression_ratio": 1.6234309623430963,
            "no_speech_prob": 0.0004868507385253906,
            "words": [
                {
                    "start": 2384.13,
                    "end": 2384.83,
                    "word": " to",
                    "probability": 0.89208984375
                },
                {
                    "start": 2384.83,
                    "end": 2385.11,
                    "word": " understand",
                    "probability": 0.73876953125
                },
                {
                    "start": 2385.11,
                    "end": 2385.37,
                    "word": " it",
                    "probability": 0.8974609375
                },
                {
                    "start": 2385.37,
                    "end": 2385.53,
                    "word": " and",
                    "probability": 0.8828125
                },
                {
                    "start": 2385.53,
                    "end": 2385.67,
                    "word": " to",
                    "probability": 0.90087890625
                },
                {
                    "start": 2385.67,
                    "end": 2385.97,
                    "word": " implement",
                    "probability": 0.74267578125
                },
                {
                    "start": 2385.97,
                    "end": 2386.33,
                    "word": " it",
                    "probability": 0.90380859375
                },
                {
                    "start": 2386.33,
                    "end": 2386.85,
                    "word": ".",
                    "probability": 0.73095703125
                }
            ]
        },
        {
            "id": 437,
            "seek": 226471,
            "start": 2386.85,
            "end": 2389.41,
            "text": " And I hope that in 10 years or less,",
            "tokens": [
                51164,
                400,
                286,
                1454,
                300,
                294,
                1266,
                924,
                420,
                1570,
                11,
                51264
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1659962800469729,
            "compression_ratio": 1.6234309623430963,
            "no_speech_prob": 0.0004868507385253906,
            "words": [
                {
                    "start": 2386.85,
                    "end": 2387.11,
                    "word": " And",
                    "probability": 0.76416015625
                },
                {
                    "start": 2387.11,
                    "end": 2387.29,
                    "word": " I",
                    "probability": 0.99169921875
                },
                {
                    "start": 2387.29,
                    "end": 2387.69,
                    "word": " hope",
                    "probability": 0.814453125
                },
                {
                    "start": 2387.69,
                    "end": 2387.93,
                    "word": " that",
                    "probability": 0.8818359375
                },
                {
                    "start": 2387.93,
                    "end": 2388.25,
                    "word": " in",
                    "probability": 0.8828125
                },
                {
                    "start": 2388.25,
                    "end": 2388.49,
                    "word": " 10",
                    "probability": 0.75
                },
                {
                    "start": 2388.49,
                    "end": 2388.87,
                    "word": " years",
                    "probability": 0.87060546875
                },
                {
                    "start": 2388.87,
                    "end": 2389.07,
                    "word": " or",
                    "probability": 0.904296875
                },
                {
                    "start": 2389.07,
                    "end": 2389.37,
                    "word": " less",
                    "probability": 0.87255859375
                },
                {
                    "start": 2389.37,
                    "end": 2389.41,
                    "word": ",",
                    "probability": 0.75048828125
                }
            ]
        },
        {
            "id": 438,
            "seek": 226471,
            "start": 2389.41,
            "end": 2391.49,
            "text": " we get to the point where crypto is nowadays,",
            "tokens": [
                51314,
                321,
                483,
                281,
                264,
                935,
                689,
                17240,
                307,
                13434,
                11,
                51364
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1659962800469729,
            "compression_ratio": 1.6234309623430963,
            "no_speech_prob": 0.0004868507385253906,
            "words": [
                {
                    "start": 2389.41,
                    "end": 2389.55,
                    "word": " we",
                    "probability": 0.89697265625
                },
                {
                    "start": 2389.55,
                    "end": 2389.75,
                    "word": " get",
                    "probability": 0.77685546875
                },
                {
                    "start": 2389.75,
                    "end": 2389.87,
                    "word": " to",
                    "probability": 0.90087890625
                },
                {
                    "start": 2389.87,
                    "end": 2390.01,
                    "word": " the",
                    "probability": 0.8134765625
                },
                {
                    "start": 2390.01,
                    "end": 2390.17,
                    "word": " point",
                    "probability": 0.859375
                },
                {
                    "start": 2390.17,
                    "end": 2390.33,
                    "word": " where",
                    "probability": 0.84521484375
                },
                {
                    "start": 2390.33,
                    "end": 2390.63,
                    "word": " crypto",
                    "probability": 0.66748046875
                },
                {
                    "start": 2390.63,
                    "end": 2390.83,
                    "word": " is",
                    "probability": 0.87451171875
                },
                {
                    "start": 2390.83,
                    "end": 2391.25,
                    "word": " nowadays",
                    "probability": 0.80419921875
                },
                {
                    "start": 2391.25,
                    "end": 2391.49,
                    "word": ",",
                    "probability": 0.556640625
                }
            ]
        },
        {
            "id": 439,
            "seek": 226471,
            "start": 2391.49,
            "end": 2394.53,
            "text": " where there are standard libraries",
            "tokens": [
                51414,
                689,
                456,
                366,
                3832,
                15148,
                51514
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1659962800469729,
            "compression_ratio": 1.6234309623430963,
            "no_speech_prob": 0.0004868507385253906,
            "words": [
                {
                    "start": 2391.49,
                    "end": 2391.69,
                    "word": " where",
                    "probability": 0.88134765625
                },
                {
                    "start": 2391.69,
                    "end": 2393.33,
                    "word": " there",
                    "probability": 0.6591796875
                },
                {
                    "start": 2393.33,
                    "end": 2393.79,
                    "word": " are",
                    "probability": 0.8935546875
                },
                {
                    "start": 2393.79,
                    "end": 2394.03,
                    "word": " standard",
                    "probability": 0.6962890625
                },
                {
                    "start": 2394.03,
                    "end": 2394.53,
                    "word": " libraries",
                    "probability": 0.8876953125
                }
            ]
        },
        {
            "id": 440,
            "seek": 226471,
            "start": 2394.53,
            "end": 2398.15,
            "text": " and you can do crypto without being an expert.",
            "tokens": [
                51564,
                293,
                291,
                393,
                360,
                17240,
                1553,
                885,
                364,
                5844,
                13,
                51714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.1659962800469729,
            "compression_ratio": 1.6234309623430963,
            "no_speech_prob": 0.0004868507385253906,
            "words": [
                {
                    "start": 2394.53,
                    "end": 2394.79,
                    "word": " and",
                    "probability": 0.67529296875
                },
                {
                    "start": 2394.79,
                    "end": 2394.97,
                    "word": " you",
                    "probability": 0.8955078125
                },
                {
                    "start": 2394.97,
                    "end": 2395.21,
                    "word": " can",
                    "probability": 0.90283203125
                },
                {
                    "start": 2395.21,
                    "end": 2395.97,
                    "word": " do",
                    "probability": 0.79541015625
                },
                {
                    "start": 2395.97,
                    "end": 2396.33,
                    "word": " crypto",
                    "probability": 0.74072265625
                },
                {
                    "start": 2396.33,
                    "end": 2396.77,
                    "word": " without",
                    "probability": 0.81201171875
                },
                {
                    "start": 2396.77,
                    "end": 2397.13,
                    "word": " being",
                    "probability": 0.876953125
                },
                {
                    "start": 2397.13,
                    "end": 2397.29,
                    "word": " an",
                    "probability": 0.90625
                },
                {
                    "start": 2397.29,
                    "end": 2397.61,
                    "word": " expert",
                    "probability": 0.80810546875
                },
                {
                    "start": 2397.61,
                    "end": 2398.15,
                    "word": ".",
                    "probability": 0.2362060546875
                }
            ]
        },
        {
            "id": 441,
            "seek": 229393,
            "start": 2398.15,
            "end": 2401.51,
            "text": " And you can rely on some other experts to do the right work",
            "tokens": [
                50364,
                400,
                291,
                393,
                10687,
                322,
                512,
                661,
                8572,
                281,
                360,
                264,
                558,
                589,
                50514
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17237102997208398,
            "compression_ratio": 1.7338129496402879,
            "no_speech_prob": 0.0025081634521484375,
            "words": [
                {
                    "start": 2398.15,
                    "end": 2398.35,
                    "word": " And",
                    "probability": 0.27197265625
                },
                {
                    "start": 2398.35,
                    "end": 2398.45,
                    "word": " you",
                    "probability": 0.85888671875
                },
                {
                    "start": 2398.45,
                    "end": 2398.59,
                    "word": " can",
                    "probability": 0.9013671875
                },
                {
                    "start": 2398.59,
                    "end": 2398.85,
                    "word": " rely",
                    "probability": 0.8369140625
                },
                {
                    "start": 2398.85,
                    "end": 2399.19,
                    "word": " on",
                    "probability": 0.90869140625
                },
                {
                    "start": 2399.19,
                    "end": 2399.71,
                    "word": " some",
                    "probability": 0.7841796875
                },
                {
                    "start": 2399.71,
                    "end": 2399.97,
                    "word": " other",
                    "probability": 0.44580078125
                },
                {
                    "start": 2399.97,
                    "end": 2400.35,
                    "word": " experts",
                    "probability": 0.740234375
                },
                {
                    "start": 2400.35,
                    "end": 2400.77,
                    "word": " to",
                    "probability": 0.888671875
                },
                {
                    "start": 2400.77,
                    "end": 2400.91,
                    "word": " do",
                    "probability": 0.7666015625
                },
                {
                    "start": 2400.91,
                    "end": 2401.03,
                    "word": " the",
                    "probability": 0.8125
                },
                {
                    "start": 2401.03,
                    "end": 2401.23,
                    "word": " right",
                    "probability": 0.8720703125
                },
                {
                    "start": 2401.23,
                    "end": 2401.51,
                    "word": " work",
                    "probability": 0.78076171875
                }
            ]
        },
        {
            "id": 442,
            "seek": 229393,
            "start": 2401.51,
            "end": 2403.83,
            "text": " so that you can be safe.",
            "tokens": [
                50564,
                370,
                300,
                291,
                393,
                312,
                3273,
                13,
                50614
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17237102997208398,
            "compression_ratio": 1.7338129496402879,
            "no_speech_prob": 0.0025081634521484375,
            "words": [
                {
                    "start": 2401.51,
                    "end": 2401.69,
                    "word": " so",
                    "probability": 0.8134765625
                },
                {
                    "start": 2401.69,
                    "end": 2401.83,
                    "word": " that",
                    "probability": 0.85693359375
                },
                {
                    "start": 2401.83,
                    "end": 2401.99,
                    "word": " you",
                    "probability": 0.89892578125
                },
                {
                    "start": 2401.99,
                    "end": 2402.37,
                    "word": " can",
                    "probability": 0.904296875
                },
                {
                    "start": 2402.37,
                    "end": 2403.03,
                    "word": " be",
                    "probability": 0.87744140625
                },
                {
                    "start": 2403.03,
                    "end": 2403.39,
                    "word": " safe",
                    "probability": 0.89306640625
                },
                {
                    "start": 2403.39,
                    "end": 2403.83,
                    "word": ".",
                    "probability": 0.974609375
                }
            ]
        },
        {
            "id": 443,
            "seek": 229393,
            "start": 2403.83,
            "end": 2406.05,
            "text": " So I hope that differential privacy reaches that point",
            "tokens": [
                50664,
                407,
                286,
                1454,
                300,
                15756,
                11427,
                14235,
                300,
                935,
                50714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17237102997208398,
            "compression_ratio": 1.7338129496402879,
            "no_speech_prob": 0.0025081634521484375,
            "words": [
                {
                    "start": 2403.83,
                    "end": 2403.89,
                    "word": " So",
                    "probability": 0.8857421875
                },
                {
                    "start": 2403.89,
                    "end": 2404.03,
                    "word": " I",
                    "probability": 0.92626953125
                },
                {
                    "start": 2404.03,
                    "end": 2404.21,
                    "word": " hope",
                    "probability": 0.8232421875
                },
                {
                    "start": 2404.21,
                    "end": 2404.29,
                    "word": " that",
                    "probability": 0.44091796875
                },
                {
                    "start": 2404.29,
                    "end": 2404.61,
                    "word": " differential",
                    "probability": 0.60595703125
                },
                {
                    "start": 2404.61,
                    "end": 2404.95,
                    "word": " privacy",
                    "probability": 0.83935546875
                },
                {
                    "start": 2404.95,
                    "end": 2405.37,
                    "word": " reaches",
                    "probability": 0.5244140625
                },
                {
                    "start": 2405.37,
                    "end": 2405.69,
                    "word": " that",
                    "probability": 0.87548828125
                },
                {
                    "start": 2405.69,
                    "end": 2406.05,
                    "word": " point",
                    "probability": 0.8662109375
                }
            ]
        },
        {
            "id": 444,
            "seek": 229393,
            "start": 2406.05,
            "end": 2408.29,
            "text": " where it becomes so commonplace",
            "tokens": [
                50764,
                689,
                309,
                3643,
                370,
                2689,
                6742,
                50814
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17237102997208398,
            "compression_ratio": 1.7338129496402879,
            "no_speech_prob": 0.0025081634521484375,
            "words": [
                {
                    "start": 2406.05,
                    "end": 2406.27,
                    "word": " where",
                    "probability": 0.86083984375
                },
                {
                    "start": 2406.27,
                    "end": 2406.49,
                    "word": " it",
                    "probability": 0.90087890625
                },
                {
                    "start": 2406.49,
                    "end": 2406.83,
                    "word": " becomes",
                    "probability": 0.81396484375
                },
                {
                    "start": 2406.83,
                    "end": 2407.45,
                    "word": " so",
                    "probability": 0.8828125
                },
                {
                    "start": 2407.45,
                    "end": 2408.29,
                    "word": " commonplace",
                    "probability": 0.8623046875
                }
            ]
        },
        {
            "id": 445,
            "seek": 229393,
            "start": 2408.29,
            "end": 2411.43,
            "text": " and a new default way of doing data analysis",
            "tokens": [
                50864,
                293,
                257,
                777,
                7576,
                636,
                295,
                884,
                1412,
                5215,
                51014
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17237102997208398,
            "compression_ratio": 1.7338129496402879,
            "no_speech_prob": 0.0025081634521484375,
            "words": [
                {
                    "start": 2408.29,
                    "end": 2408.71,
                    "word": " and",
                    "probability": 0.7890625
                },
                {
                    "start": 2408.93,
                    "end": 2409.15,
                    "word": " a new",
                    "probability": 0.82177734375
                },
                {
                    "start": 2409.61,
                    "end": 2410.23,
                    "word": " default way",
                    "probability": 0.88330078125
                },
                {
                    "start": 2410.23,
                    "end": 2410.45,
                    "word": " of",
                    "probability": 0.90576171875
                },
                {
                    "start": 2410.45,
                    "end": 2410.67,
                    "word": " doing",
                    "probability": 0.919921875
                },
                {
                    "start": 2410.67,
                    "end": 2410.91,
                    "word": " data",
                    "probability": 0.6083984375
                },
                {
                    "start": 2410.91,
                    "end": 2411.43,
                    "word": " analysis",
                    "probability": 0.78466796875
                }
            ]
        },
        {
            "id": 446,
            "seek": 229393,
            "start": 2411.43,
            "end": 2415.47,
            "text": " that is backed up by a lot of expert code",
            "tokens": [
                51064,
                300,
                307,
                20391,
                493,
                538,
                257,
                688,
                295,
                5844,
                3089,
                51214
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17237102997208398,
            "compression_ratio": 1.7338129496402879,
            "no_speech_prob": 0.0025081634521484375,
            "words": [
                {
                    "start": 2411.43,
                    "end": 2411.99,
                    "word": " that",
                    "probability": 0.6416015625
                },
                {
                    "start": 2411.99,
                    "end": 2412.15,
                    "word": " is",
                    "probability": 0.84033203125
                },
                {
                    "start": 2412.15,
                    "end": 2412.37,
                    "word": " backed",
                    "probability": 0.79833984375
                },
                {
                    "start": 2412.37,
                    "end": 2412.73,
                    "word": " up",
                    "probability": 0.86181640625
                },
                {
                    "start": 2412.73,
                    "end": 2413.13,
                    "word": " by",
                    "probability": 0.9033203125
                },
                {
                    "start": 2413.75,
                    "end": 2413.89,
                    "word": " a lot",
                    "probability": 0.888671875
                },
                {
                    "start": 2413.89,
                    "end": 2414.07,
                    "word": " of",
                    "probability": 0.90625
                },
                {
                    "start": 2414.07,
                    "end": 2414.49,
                    "word": " expert",
                    "probability": 0.76416015625
                },
                {
                    "start": 2414.49,
                    "end": 2415.47,
                    "word": " code",
                    "probability": 0.78515625
                }
            ]
        },
        {
            "id": 447,
            "seek": 229393,
            "start": 2415.47,
            "end": 2417.67,
            "text": " that has been reviewed and vetted.",
            "tokens": [
                51264,
                300,
                575,
                668,
                18429,
                293,
                371,
                46508,
                13,
                51314
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17237102997208398,
            "compression_ratio": 1.7338129496402879,
            "no_speech_prob": 0.0025081634521484375,
            "words": [
                {
                    "start": 2415.47,
                    "end": 2415.93,
                    "word": " that",
                    "probability": 0.86474609375
                },
                {
                    "start": 2415.93,
                    "end": 2416.07,
                    "word": " has",
                    "probability": 0.88427734375
                },
                {
                    "start": 2416.07,
                    "end": 2416.21,
                    "word": " been",
                    "probability": 0.82373046875
                },
                {
                    "start": 2416.21,
                    "end": 2416.61,
                    "word": " reviewed",
                    "probability": 0.68603515625
                },
                {
                    "start": 2416.61,
                    "end": 2416.87,
                    "word": " and",
                    "probability": 0.86376953125
                },
                {
                    "start": 2416.87,
                    "end": 2417.17,
                    "word": " vetted",
                    "probability": 0.842529296875
                },
                {
                    "start": 2417.17,
                    "end": 2417.67,
                    "word": ".",
                    "probability": 0.974609375
                }
            ]
        },
        {
            "id": 448,
            "seek": 229393,
            "start": 2418.33,
            "end": 2419.85,
            "text": " And just like you don't roll your own crypto,",
            "tokens": [
                51364,
                400,
                445,
                411,
                291,
                500,
                380,
                3373,
                428,
                1065,
                17240,
                11,
                51414
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17237102997208398,
            "compression_ratio": 1.7338129496402879,
            "no_speech_prob": 0.0025081634521484375,
            "words": [
                {
                    "start": 2418.33,
                    "end": 2418.43,
                    "word": " And",
                    "probability": 0.414306640625
                },
                {
                    "start": 2418.43,
                    "end": 2418.57,
                    "word": " just",
                    "probability": 0.87255859375
                },
                {
                    "start": 2418.57,
                    "end": 2418.73,
                    "word": " like",
                    "probability": 0.8681640625
                },
                {
                    "start": 2418.73,
                    "end": 2418.81,
                    "word": " you",
                    "probability": 0.51220703125
                },
                {
                    "start": 2418.81,
                    "end": 2418.93,
                    "word": " don't",
                    "probability": 0.86572265625
                },
                {
                    "start": 2418.93,
                    "end": 2419.15,
                    "word": " roll",
                    "probability": 0.72607421875
                },
                {
                    "start": 2419.15,
                    "end": 2419.27,
                    "word": " your",
                    "probability": 0.78369140625
                },
                {
                    "start": 2419.27,
                    "end": 2419.41,
                    "word": " own",
                    "probability": 0.81884765625
                },
                {
                    "start": 2419.41,
                    "end": 2419.73,
                    "word": " crypto",
                    "probability": 0.70263671875
                },
                {
                    "start": 2419.73,
                    "end": 2419.85,
                    "word": ",",
                    "probability": 0.84130859375
                }
            ]
        },
        {
            "id": 449,
            "seek": 229393,
            "start": 2419.85,
            "end": 2422.07,
            "text": " you don't do the same thing with differential privacy.",
            "tokens": [
                51464,
                291,
                500,
                380,
                360,
                264,
                912,
                551,
                365,
                15756,
                11427,
                13,
                51514
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17237102997208398,
            "compression_ratio": 1.7338129496402879,
            "no_speech_prob": 0.0025081634521484375,
            "words": [
                {
                    "start": 2419.85,
                    "end": 2419.93,
                    "word": " you",
                    "probability": 0.8916015625
                },
                {
                    "start": 2419.93,
                    "end": 2420.05,
                    "word": " don't",
                    "probability": 0.94775390625
                },
                {
                    "start": 2420.05,
                    "end": 2420.17,
                    "word": " do",
                    "probability": 0.89501953125
                },
                {
                    "start": 2420.17,
                    "end": 2420.29,
                    "word": " the",
                    "probability": 0.81396484375
                },
                {
                    "start": 2420.29,
                    "end": 2420.47,
                    "word": " same",
                    "probability": 0.81787109375
                },
                {
                    "start": 2420.47,
                    "end": 2420.63,
                    "word": " thing",
                    "probability": 0.81494140625
                },
                {
                    "start": 2420.63,
                    "end": 2420.77,
                    "word": " with",
                    "probability": 0.81689453125
                },
                {
                    "start": 2420.77,
                    "end": 2421.11,
                    "word": " differential",
                    "probability": 0.72802734375
                },
                {
                    "start": 2421.11,
                    "end": 2421.61,
                    "word": " privacy",
                    "probability": 0.8251953125
                },
                {
                    "start": 2421.61,
                    "end": 2422.07,
                    "word": ".",
                    "probability": 0.96484375
                }
            ]
        },
        {
            "id": 450,
            "seek": 229393,
            "start": 2422.07,
            "end": 2424.69,
            "text": " So the goal of our training is to really make it",
            "tokens": [
                51564,
                407,
                264,
                3387,
                295,
                527,
                3097,
                307,
                281,
                534,
                652,
                309,
                51664
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17237102997208398,
            "compression_ratio": 1.7338129496402879,
            "no_speech_prob": 0.0025081634521484375,
            "words": [
                {
                    "start": 2422.07,
                    "end": 2422.31,
                    "word": " So",
                    "probability": 0.8935546875
                },
                {
                    "start": 2422.31,
                    "end": 2422.49,
                    "word": " the",
                    "probability": 0.8076171875
                },
                {
                    "start": 2422.49,
                    "end": 2422.79,
                    "word": " goal",
                    "probability": 0.9033203125
                },
                {
                    "start": 2422.79,
                    "end": 2423.05,
                    "word": " of",
                    "probability": 0.90966796875
                },
                {
                    "start": 2423.05,
                    "end": 2423.31,
                    "word": " our",
                    "probability": 0.80078125
                },
                {
                    "start": 2423.31,
                    "end": 2423.65,
                    "word": " training",
                    "probability": 0.80517578125
                },
                {
                    "start": 2423.65,
                    "end": 2423.89,
                    "word": " is",
                    "probability": 0.9033203125
                },
                {
                    "start": 2423.89,
                    "end": 2424.03,
                    "word": " to",
                    "probability": 0.8955078125
                },
                {
                    "start": 2424.03,
                    "end": 2424.27,
                    "word": " really",
                    "probability": 0.78369140625
                },
                {
                    "start": 2424.27,
                    "end": 2424.57,
                    "word": " make",
                    "probability": 0.87353515625
                },
                {
                    "start": 2424.57,
                    "end": 2424.69,
                    "word": " it",
                    "probability": 0.90185546875
                }
            ]
        },
        {
            "id": 451,
            "seek": 229393,
            "start": 2424.69,
            "end": 2427.37,
            "text": " as accessible to everyone as possible.",
            "tokens": [
                51714,
                382,
                9515,
                281,
                1518,
                382,
                1944,
                13,
                51764
            ],
            "temperature": 0.0,
            "avg_logprob": -0.17237102997208398,
            "compression_ratio": 1.7338129496402879,
            "no_speech_prob": 0.0025081634521484375,
            "words": [
                {
                    "start": 2424.69,
                    "end": 2424.99,
                    "word": " as",
                    "probability": 0.9130859375
                },
                {
                    "start": 2424.99,
                    "end": 2425.47,
                    "word": " accessible",
                    "probability": 0.90771484375
                },
                {
                    "start": 2425.47,
                    "end": 2425.83,
                    "word": " to",
                    "probability": 0.90673828125
                },
                {
                    "start": 2425.83,
                    "end": 2426.15,
                    "word": " everyone",
                    "probability": 0.88720703125
                },
                {
                    "start": 2426.15,
                    "end": 2426.39,
                    "word": " as",
                    "probability": 0.90576171875
                },
                {
                    "start": 2426.39,
                    "end": 2426.87,
                    "word": " possible",
                    "probability": 0.8369140625
                },
                {
                    "start": 2426.87,
                    "end": 2427.37,
                    "word": ".",
                    "probability": 0.93115234375
                }
            ]
        },
        {
            "id": 452,
            "seek": 232393,
            "start": 2427.37,
            "end": 2429.35,
            "text": " And so when I post the short links,",
            "tokens": [
                50364,
                400,
                370,
                562,
                286,
                2183,
                264,
                2099,
                6123,
                11,
                50414
            ],
            "temperature": 0.0,
            "avg_logprob": -0.16794064812523,
            "compression_ratio": 1.6635514018691588,
            "no_speech_prob": 0.0016145706176757812,
            "words": [
                {
                    "start": 2427.37,
                    "end": 2427.81,
                    "word": " And",
                    "probability": 0.38720703125
                },
                {
                    "start": 2427.81,
                    "end": 2427.99,
                    "word": " so",
                    "probability": 0.85595703125
                },
                {
                    "start": 2427.99,
                    "end": 2428.17,
                    "word": " when",
                    "probability": 0.705078125
                },
                {
                    "start": 2428.17,
                    "end": 2428.29,
                    "word": " I",
                    "probability": 0.99609375
                },
                {
                    "start": 2428.29,
                    "end": 2428.63,
                    "word": " post",
                    "probability": 0.82958984375
                },
                {
                    "start": 2428.63,
                    "end": 2428.77,
                    "word": " the",
                    "probability": 0.806640625
                },
                {
                    "start": 2428.77,
                    "end": 2428.99,
                    "word": " short",
                    "probability": 0.92236328125
                },
                {
                    "start": 2428.99,
                    "end": 2429.21,
                    "word": " links",
                    "probability": 0.71240234375
                },
                {
                    "start": 2429.21,
                    "end": 2429.35,
                    "word": ",",
                    "probability": 0.88134765625
                }
            ]
        },
        {
            "id": 453,
            "seek": 232393,
            "start": 2429.35,
            "end": 2430.45,
            "text": " please go through the training.",
            "tokens": [
                50464,
                1767,
                352,
                807,
                264,
                3097,
                13,
                50514
            ],
            "temperature": 0.0,
            "avg_logprob": -0.16794064812523,
            "compression_ratio": 1.6635514018691588,
            "no_speech_prob": 0.0016145706176757812,
            "words": [
                {
                    "start": 2429.35,
                    "end": 2429.53,
                    "word": " please",
                    "probability": 0.76171875
                },
                {
                    "start": 2429.53,
                    "end": 2429.71,
                    "word": " go",
                    "probability": 0.9013671875
                },
                {
                    "start": 2429.71,
                    "end": 2429.87,
                    "word": " through",
                    "probability": 0.791015625
                },
                {
                    "start": 2429.87,
                    "end": 2430.01,
                    "word": " the",
                    "probability": 0.81005859375
                },
                {
                    "start": 2430.01,
                    "end": 2430.27,
                    "word": " training",
                    "probability": 0.81298828125
                },
                {
                    "start": 2430.27,
                    "end": 2430.45,
                    "word": ".",
                    "probability": 0.61767578125
                }
            ]
        },
        {
            "id": 454,
            "seek": 232393,
            "start": 2430.45,
            "end": 2432.25,
            "text": " If you feel like there's something missing,",
            "tokens": [
                50564,
                759,
                291,
                841,
                411,
                456,
                311,
                746,
                5361,
                11,
                50614
            ],
            "temperature": 0.0,
            "avg_logprob": -0.16794064812523,
            "compression_ratio": 1.6635514018691588,
            "no_speech_prob": 0.0016145706176757812,
            "words": [
                {
                    "start": 2430.45,
                    "end": 2430.47,
                    "word": " If",
                    "probability": 0.90283203125
                },
                {
                    "start": 2430.47,
                    "end": 2430.59,
                    "word": " you",
                    "probability": 0.896484375
                },
                {
                    "start": 2430.59,
                    "end": 2430.87,
                    "word": " feel",
                    "probability": 0.92626953125
                },
                {
                    "start": 2430.87,
                    "end": 2431.03,
                    "word": " like",
                    "probability": 0.87255859375
                },
                {
                    "start": 2431.03,
                    "end": 2431.21,
                    "word": " there's",
                    "probability": 0.917724609375
                },
                {
                    "start": 2431.21,
                    "end": 2431.47,
                    "word": " something",
                    "probability": 0.72314453125
                },
                {
                    "start": 2431.47,
                    "end": 2431.85,
                    "word": " missing",
                    "probability": 0.86669921875
                },
                {
                    "start": 2431.85,
                    "end": 2432.25,
                    "word": ",",
                    "probability": 0.9482421875
                }
            ]
        },
        {
            "id": 455,
            "seek": 232393,
            "start": 2432.25,
            "end": 2434.87,
            "text": " something's not cleared, give us that feedback.",
            "tokens": [
                50664,
                746,
                311,
                406,
                19725,
                11,
                976,
                505,
                300,
                5824,
                13,
                50714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.16794064812523,
            "compression_ratio": 1.6635514018691588,
            "no_speech_prob": 0.0016145706176757812,
            "words": [
                {
                    "start": 2432.25,
                    "end": 2432.67,
                    "word": " something's",
                    "probability": 0.578125
                },
                {
                    "start": 2432.67,
                    "end": 2432.85,
                    "word": " not",
                    "probability": 0.88427734375
                },
                {
                    "start": 2432.85,
                    "end": 2433.31,
                    "word": " cleared",
                    "probability": 0.626953125
                },
                {
                    "start": 2433.31,
                    "end": 2433.47,
                    "word": ",",
                    "probability": 0.93212890625
                },
                {
                    "start": 2433.47,
                    "end": 2433.71,
                    "word": " give",
                    "probability": 0.3486328125
                },
                {
                    "start": 2433.71,
                    "end": 2433.89,
                    "word": " us",
                    "probability": 0.90625
                },
                {
                    "start": 2433.89,
                    "end": 2434.09,
                    "word": " that",
                    "probability": 0.87548828125
                },
                {
                    "start": 2434.09,
                    "end": 2434.43,
                    "word": " feedback",
                    "probability": 0.849609375
                },
                {
                    "start": 2434.43,
                    "end": 2434.87,
                    "word": ".",
                    "probability": 0.9833984375
                }
            ]
        },
        {
            "id": 456,
            "seek": 232393,
            "start": 2434.91,
            "end": 2436.91,
            "text": " Our goal is to keep iterating on this training",
            "tokens": [
                50764,
                2621,
                3387,
                307,
                281,
                1066,
                17138,
                990,
                322,
                341,
                3097,
                50814
            ],
            "temperature": 0.0,
            "avg_logprob": -0.16794064812523,
            "compression_ratio": 1.6635514018691588,
            "no_speech_prob": 0.0016145706176757812,
            "words": [
                {
                    "start": 2434.91,
                    "end": 2435.03,
                    "word": " Our",
                    "probability": 0.8994140625
                },
                {
                    "start": 2435.03,
                    "end": 2435.17,
                    "word": " goal",
                    "probability": 0.912109375
                },
                {
                    "start": 2435.17,
                    "end": 2435.31,
                    "word": " is",
                    "probability": 0.9052734375
                },
                {
                    "start": 2435.31,
                    "end": 2435.43,
                    "word": " to",
                    "probability": 0.90625
                },
                {
                    "start": 2435.43,
                    "end": 2435.61,
                    "word": " keep",
                    "probability": 0.80126953125
                },
                {
                    "start": 2435.61,
                    "end": 2436.15,
                    "word": " iterating",
                    "probability": 0.91748046875
                },
                {
                    "start": 2436.15,
                    "end": 2436.43,
                    "word": " on",
                    "probability": 0.9111328125
                },
                {
                    "start": 2436.43,
                    "end": 2436.61,
                    "word": " this",
                    "probability": 0.880859375
                },
                {
                    "start": 2436.61,
                    "end": 2436.91,
                    "word": " training",
                    "probability": 0.81298828125
                }
            ]
        },
        {
            "id": 457,
            "seek": 232393,
            "start": 2436.91,
            "end": 2440.73,
            "text": " and to make it as easy to implement as possible",
            "tokens": [
                50864,
                293,
                281,
                652,
                309,
                382,
                1858,
                281,
                4445,
                382,
                1944,
                51014
            ],
            "temperature": 0.0,
            "avg_logprob": -0.16794064812523,
            "compression_ratio": 1.6635514018691588,
            "no_speech_prob": 0.0016145706176757812,
            "words": [
                {
                    "start": 2436.91,
                    "end": 2437.15,
                    "word": " and",
                    "probability": 0.84521484375
                },
                {
                    "start": 2437.15,
                    "end": 2437.29,
                    "word": " to",
                    "probability": 0.89794921875
                },
                {
                    "start": 2437.29,
                    "end": 2437.67,
                    "word": " make",
                    "probability": 0.873046875
                },
                {
                    "start": 2437.67,
                    "end": 2437.79,
                    "word": " it",
                    "probability": 0.9052734375
                },
                {
                    "start": 2437.79,
                    "end": 2437.99,
                    "word": " as",
                    "probability": 0.9052734375
                },
                {
                    "start": 2437.99,
                    "end": 2438.67,
                    "word": " easy",
                    "probability": 0.7470703125
                },
                {
                    "start": 2438.67,
                    "end": 2438.91,
                    "word": " to",
                    "probability": 0.90380859375
                },
                {
                    "start": 2438.91,
                    "end": 2439.33,
                    "word": " implement",
                    "probability": 0.7685546875
                },
                {
                    "start": 2439.33,
                    "end": 2440.31,
                    "word": " as",
                    "probability": 0.8994140625
                },
                {
                    "start": 2440.31,
                    "end": 2440.73,
                    "word": " possible",
                    "probability": 0.84033203125
                }
            ]
        },
        {
            "id": 458,
            "seek": 232393,
            "start": 2440.73,
            "end": 2442.79,
            "text": " for everyone across the board.",
            "tokens": [
                51064,
                337,
                1518,
                2108,
                264,
                3150,
                13,
                51114
            ],
            "temperature": 0.0,
            "avg_logprob": -0.16794064812523,
            "compression_ratio": 1.6635514018691588,
            "no_speech_prob": 0.0016145706176757812,
            "words": [
                {
                    "start": 2440.73,
                    "end": 2441.03,
                    "word": " for",
                    "probability": 0.89208984375
                },
                {
                    "start": 2441.03,
                    "end": 2441.51,
                    "word": " everyone",
                    "probability": 0.865234375
                },
                {
                    "start": 2441.51,
                    "end": 2441.91,
                    "word": " across",
                    "probability": 0.7021484375
                },
                {
                    "start": 2441.91,
                    "end": 2442.09,
                    "word": " the",
                    "probability": 0.81396484375
                },
                {
                    "start": 2442.09,
                    "end": 2442.35,
                    "word": " board",
                    "probability": 0.86181640625
                },
                {
                    "start": 2442.35,
                    "end": 2442.79,
                    "word": ".",
                    "probability": 0.9970703125
                }
            ]
        },
        {
            "id": 459,
            "seek": 232393,
            "start": 2444.53,
            "end": 2445.77,
            "text": " Great. All right.",
            "tokens": [
                51164,
                3769,
                13,
                1057,
                558,
                13,
                51264
            ],
            "temperature": 0.0,
            "avg_logprob": -0.16794064812523,
            "compression_ratio": 1.6635514018691588,
            "no_speech_prob": 0.0016145706176757812,
            "words": [
                {
                    "start": 2444.53,
                    "end": 2444.97,
                    "word": " Great",
                    "probability": 0.3154296875
                },
                {
                    "start": 2444.97,
                    "end": 2445.35,
                    "word": ".",
                    "probability": 0.9013671875
                },
                {
                    "start": 2445.35,
                    "end": 2445.61,
                    "word": " All",
                    "probability": 0.7333984375
                },
                {
                    "start": 2445.61,
                    "end": 2445.75,
                    "word": " right",
                    "probability": 0.88232421875
                },
                {
                    "start": 2445.75,
                    "end": 2445.77,
                    "word": ".",
                    "probability": 0.6962890625
                }
            ]
        },
        {
            "id": 460,
            "seek": 232393,
            "start": 2445.77,
            "end": 2448.15,
            "text": " Well, I think that's all the questions that we've got right now",
            "tokens": [
                51314,
                1042,
                11,
                286,
                519,
                300,
                311,
                439,
                264,
                1651,
                300,
                321,
                600,
                658,
                558,
                586,
                51364
            ],
            "temperature": 0.0,
            "avg_logprob": -0.16794064812523,
            "compression_ratio": 1.6635514018691588,
            "no_speech_prob": 0.0016145706176757812,
            "words": [
                {
                    "start": 2445.77,
                    "end": 2445.89,
                    "word": " Well",
                    "probability": 0.7314453125
                },
                {
                    "start": 2445.89,
                    "end": 2445.89,
                    "word": ",",
                    "probability": 0.93505859375
                },
                {
                    "start": 2445.89,
                    "end": 2445.99,
                    "word": " I",
                    "probability": 0.998046875
                },
                {
                    "start": 2445.99,
                    "end": 2446.11,
                    "word": " think",
                    "probability": 0.8076171875
                },
                {
                    "start": 2446.11,
                    "end": 2446.35,
                    "word": " that's",
                    "probability": 0.94140625
                },
                {
                    "start": 2446.35,
                    "end": 2446.47,
                    "word": " all",
                    "probability": 0.89990234375
                },
                {
                    "start": 2446.47,
                    "end": 2446.59,
                    "word": " the",
                    "probability": 0.818359375
                },
                {
                    "start": 2446.59,
                    "end": 2446.93,
                    "word": " questions",
                    "probability": 0.904296875
                },
                {
                    "start": 2446.93,
                    "end": 2447.15,
                    "word": " that",
                    "probability": 0.87060546875
                },
                {
                    "start": 2447.15,
                    "end": 2447.35,
                    "word": " we've",
                    "probability": 0.92578125
                },
                {
                    "start": 2447.35,
                    "end": 2447.57,
                    "word": " got",
                    "probability": 0.88720703125
                },
                {
                    "start": 2447.57,
                    "end": 2447.77,
                    "word": " right",
                    "probability": 0.87109375
                },
                {
                    "start": 2447.77,
                    "end": 2448.15,
                    "word": " now",
                    "probability": 0.8974609375
                }
            ]
        },
        {
            "id": 461,
            "seek": 232393,
            "start": 2448.15,
            "end": 2449.67,
            "text": " for the time being.",
            "tokens": [
                51414,
                337,
                264,
                565,
                885,
                13,
                51464
            ],
            "temperature": 0.0,
            "avg_logprob": -0.16794064812523,
            "compression_ratio": 1.6635514018691588,
            "no_speech_prob": 0.0016145706176757812,
            "words": [
                {
                    "start": 2448.15,
                    "end": 2448.73,
                    "word": " for",
                    "probability": 0.8251953125
                },
                {
                    "start": 2448.73,
                    "end": 2448.85,
                    "word": " the",
                    "probability": 0.8115234375
                },
                {
                    "start": 2448.85,
                    "end": 2449.07,
                    "word": " time",
                    "probability": 0.80908203125
                },
                {
                    "start": 2449.07,
                    "end": 2449.43,
                    "word": " being",
                    "probability": 0.87646484375
                },
                {
                    "start": 2449.43,
                    "end": 2449.67,
                    "word": ".",
                    "probability": 0.60693359375
                }
            ]
        },
        {
            "id": 462,
            "seek": 232393,
            "start": 2449.67,
            "end": 2451.43,
            "text": " And I will unleash you on the Discord",
            "tokens": [
                51514,
                400,
                286,
                486,
                49814,
                291,
                322,
                264,
                32623,
                51564
            ],
            "temperature": 0.0,
            "avg_logprob": -0.16794064812523,
            "compression_ratio": 1.6635514018691588,
            "no_speech_prob": 0.0016145706176757812,
            "words": [
                {
                    "start": 2449.67,
                    "end": 2449.85,
                    "word": " And",
                    "probability": 0.80224609375
                },
                {
                    "start": 2449.85,
                    "end": 2450.07,
                    "word": " I",
                    "probability": 0.95654296875
                },
                {
                    "start": 2450.07,
                    "end": 2450.27,
                    "word": " will",
                    "probability": 0.80224609375
                },
                {
                    "start": 2450.27,
                    "end": 2450.65,
                    "word": " unleash",
                    "probability": 0.79052734375
                },
                {
                    "start": 2450.65,
                    "end": 2450.81,
                    "word": " you",
                    "probability": 0.89111328125
                },
                {
                    "start": 2450.81,
                    "end": 2450.95,
                    "word": " on",
                    "probability": 0.90283203125
                },
                {
                    "start": 2450.95,
                    "end": 2451.09,
                    "word": " the",
                    "probability": 0.80517578125
                },
                {
                    "start": 2451.09,
                    "end": 2451.43,
                    "word": " Discord",
                    "probability": 0.6669921875
                }
            ]
        },
        {
            "id": 463,
            "seek": 232393,
            "start": 2451.43,
            "end": 2453.99,
            "text": " to answer any further questions anybody may have.",
            "tokens": [
                51614,
                281,
                1867,
                604,
                3052,
                1651,
                4472,
                815,
                362,
                13,
                51664
            ],
            "temperature": 0.0,
            "avg_logprob": -0.16794064812523,
            "compression_ratio": 1.6635514018691588,
            "no_speech_prob": 0.0016145706176757812,
            "words": [
                {
                    "start": 2451.43,
                    "end": 2451.69,
                    "word": " to",
                    "probability": 0.89794921875
                },
                {
                    "start": 2451.69,
                    "end": 2451.89,
                    "word": " answer",
                    "probability": 0.88134765625
                },
                {
                    "start": 2451.89,
                    "end": 2452.07,
                    "word": " any",
                    "probability": 0.81298828125
                },
                {
                    "start": 2452.07,
                    "end": 2452.29,
                    "word": " further",
                    "probability": 0.84375
                },
                {
                    "start": 2452.29,
                    "end": 2452.73,
                    "word": " questions",
                    "probability": 0.90283203125
                },
                {
                    "start": 2452.73,
                    "end": 2453.03,
                    "word": " anybody",
                    "probability": 0.7099609375
                },
                {
                    "start": 2453.03,
                    "end": 2453.37,
                    "word": " may",
                    "probability": 0.89404296875
                },
                {
                    "start": 2453.37,
                    "end": 2453.73,
                    "word": " have",
                    "probability": 0.88037109375
                },
                {
                    "start": 2453.73,
                    "end": 2453.99,
                    "word": ".",
                    "probability": 0.9970703125
                }
            ]
        },
        {
            "id": 464,
            "seek": 232393,
            "start": 2453.99,
            "end": 2456.59,
            "text": " Thank you all so much for providing this valuable resource",
            "tokens": [
                51714,
                1044,
                291,
                439,
                370,
                709,
                337,
                6530,
                341,
                8263,
                7684,
                51814
            ],
            "temperature": 0.0,
            "avg_logprob": -0.16794064812523,
            "compression_ratio": 1.6635514018691588,
            "no_speech_prob": 0.0016145706176757812,
            "words": [
                {
                    "start": 2453.99,
                    "end": 2454.27,
                    "word": " Thank",
                    "probability": 0.8935546875
                },
                {
                    "start": 2454.27,
                    "end": 2454.43,
                    "word": " you",
                    "probability": 0.89404296875
                },
                {
                    "start": 2454.43,
                    "end": 2454.63,
                    "word": " all",
                    "probability": 0.890625
                },
                {
                    "start": 2454.63,
                    "end": 2454.77,
                    "word": " so",
                    "probability": 0.90771484375
                },
                {
                    "start": 2454.77,
                    "end": 2454.93,
                    "word": " much",
                    "probability": 0.8212890625
                },
                {
                    "start": 2454.93,
                    "end": 2455.13,
                    "word": " for",
                    "probability": 0.9013671875
                },
                {
                    "start": 2455.13,
                    "end": 2455.51,
                    "word": " providing",
                    "probability": 0.822265625
                },
                {
                    "start": 2455.51,
                    "end": 2455.87,
                    "word": " this",
                    "probability": 0.87353515625
                },
                {
                    "start": 2455.87,
                    "end": 2456.15,
                    "word": " valuable",
                    "probability": 0.75048828125
                },
                {
                    "start": 2456.15,
                    "end": 2456.59,
                    "word": " resource",
                    "probability": 0.81982421875
                }
            ]
        },
        {
            "id": 465,
            "seek": 233055,
            "start": 2457.37,
            "end": 2458.79,
            "text": " and for having this terrific talk.",
            "tokens": [
                50364,
                293,
                337,
                1419,
                341,
                20899,
                751,
                13,
                50414
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20930232766062715,
            "compression_ratio": 1.1711711711711712,
            "no_speech_prob": 0.004550933837890625,
            "words": [
                {
                    "start": 2457.37,
                    "end": 2457.45,
                    "word": " and",
                    "probability": 0.002361297607421875
                },
                {
                    "start": 2457.45,
                    "end": 2457.45,
                    "word": " for",
                    "probability": 0.0574951171875
                },
                {
                    "start": 2457.45,
                    "end": 2457.45,
                    "word": " having",
                    "probability": 0.1588134765625
                },
                {
                    "start": 2457.45,
                    "end": 2457.59,
                    "word": " this",
                    "probability": 0.708984375
                },
                {
                    "start": 2457.59,
                    "end": 2457.97,
                    "word": " terrific",
                    "probability": 0.83642578125
                },
                {
                    "start": 2457.97,
                    "end": 2458.43,
                    "word": " talk",
                    "probability": 0.82275390625
                },
                {
                    "start": 2458.43,
                    "end": 2458.79,
                    "word": ".",
                    "probability": 0.865234375
                }
            ]
        },
        {
            "id": 466,
            "seek": 233055,
            "start": 2459.29,
            "end": 2461.13,
            "text": " Have a terrific rest of your DEF CON.",
            "tokens": [
                50464,
                3560,
                257,
                20899,
                1472,
                295,
                428,
                10113,
                37,
                16596,
                13,
                50514
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20930232766062715,
            "compression_ratio": 1.1711711711711712,
            "no_speech_prob": 0.004550933837890625,
            "words": [
                {
                    "start": 2459.29,
                    "end": 2459.51,
                    "word": " Have",
                    "probability": 0.861328125
                },
                {
                    "start": 2459.61,
                    "end": 2459.81,
                    "word": " a terrific",
                    "probability": 0.7705078125
                },
                {
                    "start": 2459.81,
                    "end": 2460.09,
                    "word": " rest",
                    "probability": 0.8681640625
                },
                {
                    "start": 2460.09,
                    "end": 2460.21,
                    "word": " of",
                    "probability": 0.89892578125
                },
                {
                    "start": 2460.21,
                    "end": 2460.31,
                    "word": " your",
                    "probability": 0.77001953125
                },
                {
                    "start": 2460.31,
                    "end": 2460.49,
                    "word": " DEF",
                    "probability": 0.811767578125
                },
                {
                    "start": 2460.49,
                    "end": 2460.77,
                    "word": " CON",
                    "probability": 0.9140625
                },
                {
                    "start": 2460.77,
                    "end": 2461.13,
                    "word": ".",
                    "probability": 0.3193359375
                }
            ]
        },
        {
            "id": 467,
            "seek": 233055,
            "start": 2461.21,
            "end": 2463.21,
            "text": " And our next talk will start at 1 p.m.",
            "tokens": [
                50564,
                400,
                527,
                958,
                751,
                486,
                722,
                412,
                502,
                280,
                13,
                76,
                13,
                50614
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20930232766062715,
            "compression_ratio": 1.1711711711711712,
            "no_speech_prob": 0.004550933837890625,
            "words": [
                {
                    "start": 2461.21,
                    "end": 2461.35,
                    "word": " And",
                    "probability": 0.61962890625
                },
                {
                    "start": 2461.35,
                    "end": 2461.47,
                    "word": " our",
                    "probability": 0.81298828125
                },
                {
                    "start": 2461.47,
                    "end": 2461.67,
                    "word": " next",
                    "probability": 0.865234375
                },
                {
                    "start": 2461.67,
                    "end": 2461.91,
                    "word": " talk",
                    "probability": 0.82080078125
                },
                {
                    "start": 2461.91,
                    "end": 2462.05,
                    "word": " will",
                    "probability": 0.7724609375
                },
                {
                    "start": 2462.05,
                    "end": 2462.23,
                    "word": " start",
                    "probability": 0.8662109375
                },
                {
                    "start": 2462.23,
                    "end": 2462.35,
                    "word": " at",
                    "probability": 0.90283203125
                },
                {
                    "start": 2462.35,
                    "end": 2462.51,
                    "word": " 1",
                    "probability": 0.9306640625
                },
                {
                    "start": 2462.51,
                    "end": 2462.65,
                    "word": " p",
                    "probability": 0.411865234375
                },
                {
                    "start": 2462.65,
                    "end": 2462.85,
                    "word": ".m",
                    "probability": 0.995849609375
                },
                {
                    "start": 2462.85,
                    "end": 2463.21,
                    "word": ".",
                    "probability": 0.91845703125
                }
            ]
        },
        {
            "id": 468,
            "seek": 233055,
            "start": 2463.45,
            "end": 2463.97,
            "text": " Thank you so much.",
            "tokens": [
                50664,
                1044,
                291,
                370,
                709,
                13,
                50714
            ],
            "temperature": 0.0,
            "avg_logprob": -0.20930232766062715,
            "compression_ratio": 1.1711711711711712,
            "no_speech_prob": 0.004550933837890625,
            "words": [
                {
                    "start": 2463.45,
                    "end": 2463.81,
                    "word": " Thank",
                    "probability": 0.26123046875
                },
                {
                    "start": 2463.81,
                    "end": 2463.97,
                    "word": " you",
                    "probability": 0.89892578125
                },
                {
                    "start": 2463.97,
                    "end": 2463.97,
                    "word": " so",
                    "probability": 0.67529296875
                },
                {
                    "start": 2463.97,
                    "end": 2463.97,
                    "word": " much",
                    "probability": 0.8046875
                },
                {
                    "start": 2463.97,
                    "end": 2463.97,
                    "word": ".",
                    "probability": 0.818359375
                }
            ]
        }
    ],
    "text": " Hello, I am Miguel Guevara. And I'm Brian Gibson. I work in privacy infrastructure at Google, in particular on differential privacy. And we will talk to you about today our differential privacy. Before we start, let me play this video for you. At Google, we use data to make our services work better for you. And it's our responsibility to keep your data private, safe, and secure. That's why we continually invest in tools that protect individual privacy. Differentially private analysis is one of those tools. Many of our features, like popular times and maps, are only possible because we can learn from the activity patterns of millions of users. With differential privacy, we can protect an individual's data while still preserving the usefulness of the data at scale. Let's take a closer look at how it works. Imagine you're trying to plan a trip to the grocery store and want to know how busy it is at the moment. We use differentially private analysis on the historical and real-time data of other shoppers to show you how busy a place is. The analysis adds noise to results so you can look at how crowded a place is without identifying who was there. Differentially private analysis is what enables us to show millions of our users every day how busy a place is while protecting people's privacy with confidence. Many important research, business, and social questions can be answered by analyzing large data sets. Which is why, after years of development, Google has made its differential privacy library open source. Whether you are a researcher working on new medical treatments or a developer building an exciting app, this open source library can help you easily and affordably gain insights from data in a private, secure way. Differential privacy is just one of many ways we build privacy protections into our products. And we will continue to innovate to new technologies that keep your data private, safe, and secure. Great. Now that we've watched the video, I'll get into trying to talk a little bit more from a high level about differential privacy. The naive question that we always get is, why don't we just remove identifiers to anonymize data? Unfortunately, removing identifiers might not be good enough, as well as removing PII. There are multiple well-known data breaches resulting from publishing the quote-unquote de-identified data. One classical example involves the publication of de-identified medical records in the state of Washington. Those records could be partially joined with the voting registry using birthday and postcode. This was sufficient to identify a majority of individuals in the data set. In another classical example, as a part of a competition in 2006 that intended to improve the movie recommendation system, Netflix published movie ratings from its users. The records had the user identifiers removed, but researchers could partially join these records with IMDB databases, which were tied to identity. Such a join made possible to partially tie the full history of movie ranking to the user identifiers. This is a classical example of joining de-identified data sets and re-identifying some of the users. Finally, de-identified search history can also, in certain cases, reveal the identity of a user. So, there is a pattern here. Even if the records are published without direct identifying information, the content of those records together with some auxiliary information may make re-identification possible in certain cases. Thus, we need something better than de-identification in general. The second question that is very common to have is, well, why don't we just aggregate data? In fact, it is a very well-known technique to publish aggregated statistics with buckets of users. Usually, this involves grouping by a set of features. For example, grouping by restaurant visits, by location, and hour. This is commonly referred to as k-anonymity, since each user in the anonymized dataset is indistinguishable from at least k other users. Now, let me explain to you why we need differential privacy and how it works. Let's imagine you are attending a two-day conference and the moderator shares some insights about the audience. Here, you can see the country of origin of the participants on day one. It's aggregated, so it does not reveal anything about any of the individuals in the room, right? And it looks like the top country of origin is Italy, whereas Germany has the least participants and France and Spain are in the middle. On day two, the moderator shows the stats for the day. They look pretty similar, right? Except, someone has been paying attention and noticed that there was only one new person on day two. And when they look at the data from both days together, look at that. Can you tell where the specific individual is coming from? Yes. Germany. This is how, with two aggregated datasets, the moderator has managed to leak information about a particular participant. K-anonymity provides stronger guarantees than simple de-identification, but it is not perfect either. For example, it is problematic if I'm part of a bucket that is associated with a sensitive attribute, say a particular medical condition. It's not helpful that I'm indistinguishable from K other users having the same condition. While I cannot be associated with any particular record in the database, my sensitive attribute in that example is revealed. Here is another example. Imagine two aggregate buckets, 99 users plus me and 99 users without me. Publishing aggregate information about those two buckets is K-anonymous, but it's possible to identify me and my sensitive attributes by differencing those two buckets. At last, if the attacker can modify the database and create fake entries, there's no anonymity for my data if I'm part of a crowd of 99 fake users. Those type of attacks aren't always feasible or relevant, but they cannot be excluded in general. Is there a better way to define anonymity? I think so. Let's use differential privacy. The moderator would similarly announce that they're showing insights about the audience, but they would like to announce, I'm going to share insights about the audience, randomly changing the numbers just a little. Look at the day one stats, how these stats look compared to the real numbers. They're slightly different. So yes, it is less precise, but you still get the same insights about the statistical trends. Similarly, let's look at day two. Same thing. But now, if you compare the stats for day one and day two, you can't tell which country the single person is coming from. And you know it's not worth trying since you know the numbers were changed a little. This is the principle behind differential privacy. Differential privacy introduces the right amount of noise to the data, so we're able to provide useful statistically significant insights while ensuring that no one can tell if a particular individual's information was used in the competition in a mathematically proven way. Just to recap, and Brian will talk a little bit more about it in a bit, an algorithm is differentially private if its output doesn't change much when a single person is added to the database. This has a rigorous mathematical definition, which basically considers a pair of two neighboring data sets, calculates the output of those data sets, and defines a threshold of the probability density functions. This threshold is what defines how much the output is allowed to change, and it's called epsilon. Now, let me talk to you about some open sourcing efforts we've done at Google to share some of this technology with the world. We publish our code in an open source library so that others can reuse our work. As I mentioned before, differential privacy has a lot of implementation subtopics, and it is hard to get it right. As all of you in the crypto community probably know, people should enroll their own crypto. And this is also a really good advice for differential privacy and their implementations. Reusing a hardened implementation is generally a good idea. We open source the technology, and you can read more of the articles that were written back then. I'll now give you some insights as to how we've used the technology. You're probably aware of the Google Mobility Reports. These Mobility Reports were a response to the emerging crisis back in March around COVID-19. They provide policymakers with an idea of mobility trends at a city and state level for different categories. Before I continue going into the Mobility Reports, let me talk to you a little bit about location data at Google. Location data is an optional account feature that says where you go in your private timeline. It's off by default and can be paused or unpaused anytime. From a product perspective, it's available on Android and iOS. It helps users remember where they have been and gives users personalized information such as better restaurant recommendations, suggestions for a faster commute, etc. Users can view, edit, and delete some or all of their timeline at any time, as well as set up auto deletion after three or 18 months. More recently, we announced that auto deletion is now automatic for new accounts. And if you're curious as to how it looks like, you can access it as it's been shown on this GIF. Now, let me talk to you about the reports. The reports are the Mobility Reports. These reports were the first product that Google launched that was built with differential privacy from the ground up. The reports do not share data about any individual's locations, movements, or conflicts. The insights are created with aggregated, anonymized datasets from users who have turned on the location history setting, which is off by default. People can turn off that setting at any time and delete their history. We took the additional step of utilizing differential privacy to ensure that no individual can be identified. So how exactly are we using differential privacy? First, we produce a count of people at certain locations within a geographical area. Let's say that we count the visitors to parks in Vegas on a given day. That might give us a number of 8,238 people. Secondly, we add what we call some random noise to that number. Let's assume, for example, that we added an artificial number of 79. We add these two and get a new count of 8,317. This count is differentially private, and we can now use it to calculate the metrics that we publish. This is an example of a differentially private query that we do to produce reports like the mobility report. It's helpful for people who are familiar with the SQL environment because it uses the general syntax by only introducing new anonymization functions. In this case, an anonymization count. Generally, to produce the metrics, in the reports, we compute a baseline using differential privacy for a given weekday in the past. Second, we compute those same metrics for a given day. Let's say for Friday, March 27, 5,000 people visited parks in the same region. Then we divide that result by the first result, showing that there was a 50% reduction in visits to parks in that region. And then we repeat this process for all the metrics that we are producing. And therefore, all the metrics that we're reporting are produced with differential privacy. There are some interesting practical aspects on how we use these reports. We limit the number of times that a person is counted on each metric. For example, if for some reason I go to the grocery store 100 times in a day, I will only be counted once. This prevents specific patterns like delivery worker people from being too revealing. Second, we remove metrics that are too noisy. By using confidence intervals, we can figure out when a number has a less than 5% chance of being wrong by more than 10%. And finally, if you're curious, we published a technical blog detailing the anonymization process. Differential privacy has been around for a while, and mobility reports are not the only example out there that's using it. The US Census is using it for its 2020 census. Facebook recently released an URL data set. Apple has been using it. We released it on Chrome a while ago. And it's also open source for machine learning purposes on TensorFlow privacy. Now, Brian will talk to you a little bit more about building the infrastructure. So getting to the point where we could build a new product from the ground up with differential privacy was a long journey. But I want to take you through three really critically important pieces of that journey, which is building strong foundations, building frameworks that are familiar to potential users, and helping users to assess the privacy utility frontier. So the infrastructure that we build starts with the differentially private foundational libraries. We started these about three years ago, and our vision was to power differentially private use cases that could primarily accelerate developer velocity. A lot of times users are trying to do experiments or prototypes before actually building a project, and this ideation stage is perfect for differential privacy, because it can provide a very good way of getting trends or overall statistics, and allows people to assess whether it's even worth continuing development on a particular project. So the foundational library that we use here at Google is actually the same as the one that we open sourced and published, so that others could reuse our work. For us, it's always been important to provide transparency around our implementation. Hopefully folks have had a chance to poke around the library. It was launched in C++ several months ago, and a couple of weeks ago we also launched a Java version of this. I think the analogy we oftentimes use with differential privacy is best compared to cryptography, right? There you say, don't roll your own crypto, same thing goes for differential privacy. There are a lot of gotchas, there's a lot of difficulty involved in rolling it out correctly, and an incorrect or partial implementation of DP is in some cases worse than none at all, because you might have a gaping hole in your system that you're not aware of. So reusing a hardened implementation is just a good idea. So the second step we did was to build a SQL engine, and there were lots of languages we could have used, lots of ways we could have gotten about this, but SQL really made the most sense, right? It's the language that most developers are familiar with, you don't have to be an expert to get in and start doing an analysis on data sets, and so it just felt like the right choice. So I'm just going to give a little bit of an overview of some of the challenges involved in this. So one of the big ones is that SQL allows pretty much anything. A lot of implementations are actually Turing-complete, so enforcing differential privacy, while at the same time allowing non-experts to write SQL that they're basically familiar with is a big challenge. A lot of challenges come up around things like joins between different data sets, or cardinality changes, if you're allowed to just multiply rows endlessly, lots of privacy implications come out of that. So the basis of differential privacy is really about quantifying the effect that a single user can have on output. And part of a differentially private SQL engine really has to have the same property. So what we do is we actually split SQL statements into per-user transforms and cross-user aggregations. The per-user transforms are very generous, you can do essentially arbitrary, even Turing-complete SQL. The cross-user transforms are a restricted set of aggregation functions that are much more protected from a privacy standpoint. Finally, you really want to ensure correctness when you're talking about any type of privacy-sensitive infrastructure. And unfortunately, proving that something you've built is differentially private is really hard. It's so hard, it's actually provably undecidable. That said, as part of the foundational libraries, we employ stochastic model checking as part of regression testing. Now obviously this isn't a sufficient guarantee for differential privacy by itself, but it really quickly catches violations, both obvious and subtle, and provides a stable backdrop to more traditional privacy verification work and expert analysis. So this is just an example of a differentially private query, we'll get into this a little bit more later. It's meant to be user-centric, it highlights the fact that you're doing an anonymous query right up front. And in general, it basically looks like a normal SQL query with just a few more keywords and syntactic sugar. You see that this query includes things like count and average, and is selecting from some table in exactly the same way that you would perform normal SQL. So that was a really cursory overview of this particular implementation. You can check out the details of our SQL implementation in full in the paper that we published at PETS this year. And we'll be getting into some of the more detailed math later in the talk. So we're going to spend some time to return to the math behind differential privacy a little bit. We talked kind of about how our SQL engine was structured and a little bit about differential privacy. But going into a bit more of the details, you know, usually when a DP is introduced, it looks something like this. Of course, what we're actually saying when we say some sort of probabilistic statement about any differentially private system is that adding or removing a user from a data set doesn't change the results that much. That much actually has a really rigorous technical definition. There's two parameters, there's epsilon and delta. Epsilon, we generally think about in a sort of a Bayesian context, which is to say that supposing I have some notion of a detail of a user that you're trying to target, right? Supposing you're about 60% sure that they live in Wisconsin, and the attacker is attempting to improve that confidence from their perspective, ideally to 100%. Well, it turns out that there's a really nice translation between how much an attacker can improve their confidence in knowing something and the epsilon that you choose. So for a particular choice of epsilon, perhaps they might be able to improve only 5% their confidence if they're starting at 60%. If you really, really want to be rigorous, perhaps they're not able to really improve it, you know, in any sort of measurable way at all. And so you have this nice principled trade-off between risk and accuracy, since the epsilon has the strongest influence on sort of the quality, the accuracy, the utility of the output results. Delta is a much easier concept to understand and explain. It's basically our out to say that we basically made a differentially private engine, but occasionally, very occasionally, it's not differentially private. So delta is the probability of catastrophic failure. There are lots of ways to think about it, but the way we generally talk about it is perhaps the number of output rows that are not strictly differentially private. Now, of course, you don't know which rows those are, so there's always sort of a reasonable doubt. But this allows you to really quantify a baseline of failure modes that your output data set can have. Delta tends to be very, very small, typically on the order of 1 over n, where n is the number of users, or rows, rather. And because the nature of the noise that we use is a decaying exponential, the delta that is sort of induced by some of the math, actually can be sort of smaller than is machine representable in many cases, as we'll get into in a few minutes. So oftentimes, you don't actually have to worry about a delta, but it's sort of important enough that you should be able to reason about it. So just returning earlier, we talked about the privacy model, per-user transformation that's very generous, more restricted cross-user aggregation. And that's why we apply to SQL. Let's take kind of an example of a typical privacy-sensitive query that seems pretty innocuous. We're counting up the number of visits to a website based on browser agent. So a quick look through the logs reveals that a single user used a particular agent. And already, we've kind of re-identified an individual without doing anything more than counting statistics. So we can see the problem with kind of a non-differentially private environment already. So classically, sort of naively, one thing that one does in differential privacy is they add some kind of noise that satisfies differential privacy to counts. And so in this case, we go through all the browser agents, we adjust the counts, plus or minus, a small amount of noise dependent on epsilon. And in general, you might think, oh, we're good. Except that a browser agent can be an arbitrary string, right? So this could be a made-up thing. Essentially, from the position of an attacker, you have to assume that browser agent is an unbounded space of potentially arbitrarily linked strings. Which means that if you see a value there in your logs, then at least one user must have registered that. There are lots of ways of dealing with this. You could generate a bunch of random strings, and sort of give essentially zero noise to each of those. But you, for an unbounded space, would have to generate unboundedly many browser agents. So the way that we do it is we basically just filter and say that at least tau users need to contribute to any particular partition. It turns out that there's a direct line between tau and delta. So it's really easy to reason about what sort of the correct tau or the correct delta needs to be in terms of the existing parameters. As I said earlier, the nature of the noise is a decaying exponential. So even for 50 users or 100 users, you might be at a vanishingly small, perhaps unmeasurably small, probability that a particular partition is surfaced incorrectly. So you think, we're done, right? Well, almost. There's a little bit more nuance to this. And this is kind of one of the big strengths of this type of a privacy model in SQL is that you can enforce it. So an individual user in any particular session might be using a single browser. But across all their sessions, they may have used many browsers, unboundedly many, in fact. And from the perspective of differential privacy, we're trying to protect a user, not a browser. So if a user has used 10 different browsers and everybody else has just used one, potentially you violated kind of the epsilon guarantees of your system, if you allow that. So the way that we deal with that is with reservoir sampling of a user's contributions and then limiting those contributions. So a user might be able to contribute to 10 separate partitions or three or one or whatever seems appropriate. And again, it's another parameter that can be chosen by the user. But you can also sort of choose one based on an appropriate epsilon and delta. So again, it's just about giving an engine the flexibility to represent sort of both utility and privacy needs at the same time. And this is basically the model in a nutshell. Now, of course, if you were requiring users to write exactly the syntax every time, that's not possible. And so the new syntax is introduced with anonymization is basically just a clever rewriter that is doing all of that in the background to the AST of the SQL and allowing people to write basically SQL that they're familiar with. You see a few more keywords in here, obviously, the privacy parameters, the anonymization keyword in the aggregation calls. And of course, sort of importantly, it's really critical to differential privacy to be able to bound what a single user is able to contribute. So, you know, from the perspective of a particular partition, perhaps, you know, you only want a user to be able to contribute a single visit to a website. But perhaps for a salary, you'd want to allow them to contribute up to, you know, $1 million, $1 million, whatever seems reasonable there. Perhaps if you were contributing an average age or something, you'd want to bound people between 0 and 120. The more that you can definitely say about aggregation bounds, the more accuracy you can do. But of course, a lot of times when you're dealing with arbitrary data, machine learning, that sort of thing, you have no idea what bounds that are in the data. In fact, that's maybe the reason you're doing the analysis in the first place is to learn a bit more about the data. And so one of the other contributions of this work is automatic bounds determination. You want to be able to sort of get the best possible bounding on your data without paying for it very much. Once again, finding out global bounds that are differentially private on arbitrary functions is undecidable. So, you know, you can just sort of give up out of the gate and just sort of go to the next best thing, which is a good approximation. And so what we use when bounds are not provided is logarithmic histograms. Basically, you have partitions of increasing size. Doubling, for instance, allows you to get a reasonably good upper bound that might be off by a factor of two, which in terms of differential privacy actually doesn't cost you very much in terms of utility, but still gives you a really solid understanding of, you know, sort of the worst case upper bound. And what's really amazing about this is that, you know, even with our epsilon having to be split between the exploration bounds of the histogram and the actual computation, the accuracy is often still better than manually determined bounds. Again, just because you're oftentimes talking about worst case scenarios when you're determining bounds in a computation. And this is finding them sort of data dependent bounds. And typically, you know, you might require maybe 100 elements or fewer, depending on the nature of the calculation. Now, let me go back to our open sourcing availability and some action items for you. First, we would like to take part of our differentially private training that we're piloting with this audience. As a first step, please take our survey. And it only takes five minutes. It tries to figure out how much have you understood about differential privacy before you jump into the next topics. Then you can check another introductory video on differential privacy in that link. Third, you can try out some of our code labs. The link that's pointing out there points to a Go pipeline that you can build to produce differentially private metrics for a restaurant and its busy times. Fourth, if you're curious about differential privacy and learning more, you can check out the blog that we're linking. Finally, provide us some feedback on the overall training course and the materials that we've created for you. We're really looking forward to hearing more from your end. And with that, we'll open it up for questions. All right. Thank you for this terrific talk. This was differential privacy, more important than ever in the world of COVID-19. We have Aditi Joshi, Miguel Guevara, and Royce Wilson here to chat with us and answer some questions. First, I think, Aditi, you wanted to introduce the team. Hold on one moment. I think we're having an audio problem. We'll wait for a cue. In the meantime, thanks, everybody, for your patience. This is our glitched episode. And just as a reminder, you can submit any questions you have for the team into the Q&A channel on the Discord. You can find the link there at the link from the Twitch stream. You can also go to our website. It contains links to Twitch. Excuse me, to our Discord. I have her try speaking. So that it was as useful and practical to the audience as possible. And with that, I'm going to turn it over to Miguel and Royce to introduce themselves. All right. One moment. Let me check with our team behind the scenes. Do we have audio right now for our speakers? Okay, we do. Great. Fantastic. Wonderful. Thanks for that intro, Aditi. Miguel, take it away. Great. Thank you, everyone, for coming today to our talk. I'm Miguel. I'm a product manager working on Google's privacy team. Basically, I work with Royce and other engineers on the team on developing our internal differential private infrastructure that we recently open sourced. And then I am Royce. I'm a sophomore engineer at Google. I work on privacy infrastructure. I was also the lead author behind the paper that the talk is based on. Terrific. So we have a couple of questions. I'll start off with the question that I have, just taking the liberty and privilege to do so. Have you encountered any misunderstandings in your work about how Google uses differential privacy? And what have you found to be most effective in combating any of those misunderstandings? That's an interesting question. I wouldn't say it's misunderstandings on how Google uses differential privacy. I would say it's just your own misunderstanding about what differential privacy is. A lot of the times when we launch new features and when we open source things, people approach to us trying to use it. They're very excited because they sort of think that differential privacy is this magic wand that when it touches any sort of data, it suddenly transforms it into private data. And I think that the biggest misunderstanding is people confuse differential privacy a lot with synthetic data. So they believe that differential privacy can automatically turn things into synthetic data. And I think that's the biggest misunderstanding. And one thing that's, I think, important to highlight is that our open source offering helps people produce statistics in a differential private manner. So if any of you has one of those use cases, then I think that the open sourcing offering is a good place to start. And that's the way that we use it at Google to produce a lot of these statistical information in a private manner. Great. Any input from Aditi or Royce on that particular issue? No, I think Miguel captured it. Great. And so this is kind of a related question, I guess. Can you comment on what Google services, besides location visit and busy time reporting, are implementing differential privacy today? Can you give us a sense? Yeah. So the mobility reports also use differential privacy for all of their reporting. We have a lot of features on the backend that use differential privacy. So for instance, Maps has this way of trying to cache tiles that you'll probably use when you have no data. And the way that Maps does the ranking on the backend is by using differential privacy. Popular Dishes on Maps also uses differential privacy to do the ranking of the dishes. And those are some of the user-facing features that we have. However, we use differential privacy a lot for internal analysis. And that internal analysis informs new features. Great. I'm going to use the link to the COVID mobility reports in the Discord channel so people can reference it. All the short links from your slide deck into Discord as well. Fantastic. And then just more broadly, and this is sort of a big long-term question, what do you think is next for differential privacy? That could be sort of in terms of the academic work or in terms of the applied work that you see your team going with in the next several years? Royce, do you want to take that question? Yeah. Miguel might be more suited for this one, I think. Do you want to...? Okay. I think that the biggest academic frontier is improving the performance of differential privacy. At the moment, for certain use cases, you get very big hits in terms of utility. And therefore, it's not good for certain use cases. So I really hope that within the academic community, they can come up with new ways of doing differential privacy with less utility trade-offs. I think that's the other thing that I was thinking about is just make it generally more usable. At the moment, it requires a significant level of expertise to understand it and to implement it. And I hope that in 10 years or less, we get to the point where crypto is nowadays, where there are standard libraries and you can do crypto without being an expert. And you can rely on some other experts to do the right work so that you can be safe. So I hope that differential privacy reaches that point where it becomes so commonplace and a new default way of doing data analysis that is backed up by a lot of expert code that has been reviewed and vetted. And just like you don't roll your own crypto, you don't do the same thing with differential privacy. So the goal of our training is to really make it as accessible to everyone as possible. And so when I post the short links, please go through the training. If you feel like there's something missing, something's not cleared, give us that feedback. Our goal is to keep iterating on this training and to make it as easy to implement as possible for everyone across the board. Great. All right. Well, I think that's all the questions that we've got right now for the time being. And I will unleash you on the Discord to answer any further questions anybody may have. Thank you all so much for providing this valuable resource and for having this terrific talk. Have a terrific rest of your DEF CON. And our next talk will start at 1 p.m. Thank you so much."
}