{
  "generated_from": "dataset",
  "selection_rule": "Only samples listed in dataset/*/train.txt and backed by both JSON metadata and MP4 media are counted.",
  "datasets": {
    "TalkVid": {
      "name": "TalkVid",
      "split_basis": "train.txt",
      "train_list_count": 1017,
      "meta_json_count": 4091,
      "inventory_clip_count": 2347,
      "inventory_meta_without_mp4_count": 1744,
      "usable_clip_count": 1017,
      "train_missing_mp4_count": 0,
      "train_missing_meta_count": 0,
      "total_duration_seconds": 20042.96,
      "total_duration_human": "5.57 h",
      "duration_stats_seconds": {
        "mean": 19.708,
        "median": 19.12,
        "min": 6.4,
        "max": 39.92
      },
      "unique_speakers": 234,
      "clips_per_speaker": {
        "mean": 4.35,
        "median": 2.5,
        "max": 35
      },
      "speaker_count_method": "Grouped by the source video ID after stripping the clipxxx prefix. This matches the train split organization and avoids counting the same identity multiple times across duplicated clip prefixes.",
      "resolution": [
        "1080x1440",
        "1080x1920",
        "1280x720",
        "1904x1080",
        "1920x1080",
        "1920x972"
      ],
      "fps": [
        25.0
      ],
      "frame_count_stats": {
        "mean": 492.7,
        "median": 478,
        "min": 160,
        "max": 998
      },
      "face_box_stats": {
        "mean_width": 314.91,
        "mean_height": 416.81,
        "median_width": 305,
        "median_height": 407,
        "mean_area_ratio_percent": 6.72,
        "median_area_ratio_percent": 6.06
      },
      "modalities": {
        "train_mp4_clips": 1017,
        "train_wav_clips": 1017,
        "train_wav_16k_clips": 1017,
        "inventory_wav_clips": 2347,
        "inventory_wav_16k_clips": 2287,
        "original_wav_sample_rates": [
          16000
        ],
        "original_wav_channels": [
          1
        ],
        "original_wav_bits": [
          16
        ],
        "wav_16k_sample_rates": [
          16000
        ]
      },
      "metadata_schema": [
        "face_list",
        "face_size",
        "frames",
        "isvalid",
        "landmark_list",
        "mp4_path",
        "video_size",
        "wav_path"
      ],
      "format_description": {
        "root": "dataset/TalkVid",
        "meta_dir": "dataset/TalkVid/meta",
        "media_dir": "dataset/TalkVid/video_audio_clip_root",
        "split_file": "dataset/TalkVid/train.txt",
        "pairing_rule": "One JSON metadata file and one MP4 clip share the same stem; each clip also has paired WAV audio.",
        "json_fields": [
          "face_list",
          "face_size",
          "frames",
          "isvalid",
          "landmark_list",
          "mp4_path",
          "video_size",
          "wav_path"
        ],
        "json_video_size_order": "[height, width]",
        "json_face_size_order": "[height, width]",
        "annotations": [
          "per-frame face bounding boxes",
          "per-frame facial landmarks",
          "frame count",
          "paired audio/video paths",
          "face crop size",
          "validity flag"
        ]
      },
      "speaker_examples": {
        "-NrKnqFLa0A": "clip000_-NrKnqFLa0A_002_186.920_391.608",
        "-XrkF8mnLZw": "clip001_-XrkF8mnLZw_003_156.140_219.980",
        "03QhX1UHUxg": "clip000_03QhX1UHUxg_002_64.867_121.767",
        "0_BO2MU1TwA": "clip002_0_BO2MU1TwA_001_0.100_101.120",
        "3sDxwS2WWZw": "clip000_3sDxwS2WWZw_001_0.100_34.150",
        "4E6ZsYY8h3I": "clip000_4E6ZsYY8h3I_001_5.700_22.140",
        "4ElCMYQWsf8": "clip000_4ElCMYQWsf8_004_184.633_194.300",
        "4PWeU2TXp90": "clip000_4PWeU2TXp90_003_47.667_59.767",
        "4R1yqW6RFck": "clip000_4R1yqW6RFck_001_0.100_413.280",
        "4XnPcKp1Bfc": "clip000_4XnPcKp1Bfc_003_182.983_201.401",
        "4ZMaNQKFAyc": "clip000_4ZMaNQKFAyc_004_70.070_77.110",
        "4gKdJ9otoK0": "clip000_4gKdJ9otoK0_001_212.128_227.094",
        "4ntUFfG_v9k": "clip000_4ntUFfG_v9k_002_341.274_352.119",
        "4xLAfQUMpOg": "clip000_4xLAfQUMpOg_001_0.100_16.133",
        "57ge0E0ow-4": "clip000_57ge0E0ow-4_002_251.367_269.900",
        "5Cl93NoOkqk": "clip000_5Cl93NoOkqk_001_0.100_10.500",
        "5LaVZT5VFbc": "clip000_5LaVZT5VFbc_001_112.800_139.833",
        "5PdEAwErUck": "clip000_5PdEAwErUck_002_459.475_671.858",
        "5ZCq5o68pSc": "clip001_5ZCq5o68pSc_005_145.100_195.667",
        "5kVyF6M_b9s": "clip000_5kVyF6M_b9s_001_0.100_6.400",
        "68izb9mPFC0": "clip000_68izb9mPFC0_004_158.400_275.167",
        "6BApkZFA_wY": "clip000_6BApkZFA_wY_001_288.180_302.900",
        "6NKO4foW8oQ": "clip000_6NKO4foW8oQ_006_471.137_479.379",
        "6QN-rwcl7dg": "clip002_6QN-rwcl7dg_019_748.733_824.300",
        "6_B50mCayJo": "clip000_6_B50mCayJo_005_114.933_124.733",
        "6aHpgGKYgUQ": "clip000_6aHpgGKYgUQ_003_57.900_85.233",
        "6cVIotSKeE0": "clip021_6cVIotSKeE0_002_22.800_850.733",
        "6eECcd2ziCs": "clip000_6eECcd2ziCs_017_564.867_578.533",
        "7AFz0nh64G8": "clip001_7AFz0nh64G8_006_571.070_637.604",
        "7FXDEu93syU": "clip001_7FXDEu93syU_001_0.100_40.367",
        "7G0Zr6nZyIQ": "clip000_7G0Zr6nZyIQ_003_129.863_137.437",
        "7GYbPu043so": "clip000_7GYbPu043so_001_13.580_28.500",
        "7PBhGkUe7bU": "clip000_7PBhGkUe7bU_001_0.100_20.367",
        "7Rs_7Yw4pME": "clip000_7Rs_7Yw4pME_002_34.867_41.550",
        "7kBOdAsVV1Y": "clip000_7kBOdAsVV1Y_002_114.217_126.017",
        "7ncNzI43Jc0": "clip000_7ncNzI43Jc0_001_0.100_8.842",
        "7puTvx0z_YE": "clip000_7puTvx0z_YE_001_149.740_159.860",
        "7raLkDG7B8A": "clip000_7raLkDG7B8A_005_52.102_65.616",
        "7rgRKKQSLHI": "clip000_7rgRKKQSLHI_002_32.317_54.358",
        "7yWr0kSPh38": "clip000_7yWr0kSPh38_013_681.714_689.622",
        "84t5voi3TbE": "clip000_84t5voi3TbE_002_29.740_37.380",
        "86sBla6aOuU": "clip000_86sBla6aOuU_013_277.210_284.918",
        "8Fbq82NkG5I": "clip000_8Fbq82NkG5I_002_171.633_179.567",
        "8GERUZ3Tet8": "clip000_8GERUZ3Tet8_001_28.700_35.733",
        "8KrRS89GTYE": "clip000_8KrRS89GTYE_003_89.667_259.400",
        "8UVwtPwYIgg": "clip000_8UVwtPwYIgg_009_197.867_204.667",
        "8XHTT-TQkS4": "clip000_8XHTT-TQkS4_001_0.100_307.467",
        "8pWcI_o7HTk": "clip001_8pWcI_o7HTk_001_0.100_47.531",
        "8ybKJnessOc": "clip000_8ybKJnessOc_001_5.733_47.900",
        "9CRI6CMkMok": "clip000_9CRI6CMkMok_004_114.700_152.933",
        "9LOXYkEnZ5M": "clip000_9LOXYkEnZ5M_001_0.100_12.140",
        "9OCJ7r9Gifw": "clip000_9OCJ7r9Gifw_003_111.567_121.500",
        "9W_UhoiU6ps": "clip000_9W_UhoiU6ps_001_0.700_16.900",
        "9eOGgAKXm4Y": "clip000_9eOGgAKXm4Y_001_6.156_27.311",
        "9foJ-yMWQrs": "clip000_9foJ-yMWQrs_013_496.433_506.400",
        "9fxbgcZBPRE": "clip000_9fxbgcZBPRE_013_1007.073_1015.882",
        "9sbdU7ZNZsk": "clip000_9sbdU7ZNZsk_020_457.100_464.933",
        "A-Uvj0ukOWM": "clip000_A-Uvj0ukOWM_002_23.667_30.667",
        "A8fi4PhdCoM": "clip000_A8fi4PhdCoM_001_9.100_467.300",
        "AMKi1nerKvY": "clip000_AMKi1nerKvY_001_155.055_169.736",
        "A_wcaOFAIQ8": "clip000_A_wcaOFAIQ8_002_325.333_423.433",
        "B6Lgrc7VQQY": "clip000_B6Lgrc7VQQY_014_526.300_534.300",
        "B84RIA4cK18": "clip000_B84RIA4cK18_003_296.100_304.000",
        "BAfPKtLFP4M": "clip000_BAfPKtLFP4M_003_32.867_79.200",
        "BEldvvU4Pcw": "clip000_BEldvvU4Pcw_003_166.216_243.593",
        "BR9Od4MK65g": "clip000_BR9Od4MK65g_005_167.620_174.580",
        "BWTNrM9gHvc": "clip006_BWTNrM9gHvc_001_1162.100_1487.800",
        "BvdzeI0zsUk": "clip000_BvdzeI0zsUk_001_0.100_6.533",
        "CI6Wx67u5tU": "clip001_CI6Wx67u5tU_002_725.533_798.800",
        "CJxTHYRpy6w": "clip000_CJxTHYRpy6w_005_1285.000_1305.920",
        "CbXcqAXxqdo": "clip000_CbXcqAXxqdo_004_415.532_433.233",
        "DWiGs8QEoXM": "clip000_DWiGs8QEoXM_001_9.993_125.709",
        "DfHNJv2XyQ4": "clip000_DfHNJv2XyQ4_001_255.467_262.933",
        "Duazl_Dx2Ac": "clip000_Duazl_Dx2Ac_002_119.033_132.233",
        "Dv-cx2KrO58": "clip001_Dv-cx2KrO58_014_320.900_361.900",
        "E2oNkLXuc7k": "clip000_E2oNkLXuc7k_002_34.860_63.060",
        "E5bc1hRrchM": "clip000_E5bc1hRrchM_001_0.100_7.533",
        "EJsQ1QVQgpc": "clip000_EJsQ1QVQgpc_001_34.833_46.367",
        "E_W38li04s0": "clip000_E_W38li04s0_001_0.100_29.867",
        "EfbNauJZEzc": "clip000_EfbNauJZEzc_001_4.533_55.833",
        "F8gL-EEU0lE": "clip000_F8gL-EEU0lE_001_179.734_204.199",
        "FQkYwEY3EkU": "clip000_FQkYwEY3EkU_001_744.567_752.667",
        "FWyYNBsYNOw": "clip000_FWyYNBsYNOw_001_0.100_18.200",
        "FXHuTFwM_PY": "clip000_FXHuTFwM_PY_001_0.100_8.275",
        "Hi3cylHtYAk": "clip000_Hi3cylHtYAk_001_158.525_183.750",
        "InSZLAKTp4c": "clip001_InSZLAKTp4c_003_40.067_232.533",
        "Ip3tfs9nBYE": "clip001_Ip3tfs9nBYE_001_0.100_72.467",
        "JM3geQUqlaY": "clip000_JM3geQUqlaY_019_771.033_847.000",
        "JkFDKkTeixE": "clip000_JkFDKkTeixE_008_123.457_130.297",
        "Jo0Y2Z2P2nk": "clip001_Jo0Y2Z2P2nk_002_27.000_678.500",
        "JyjHWz0xdAI": "clip001_JyjHWz0xdAI_036_660.600_698.567",
        "K0Dw3LIJGJw": "clip000_K0Dw3LIJGJw_002_183.950_191.258",
        "KEoiCXMgRHU": "clip000_KEoiCXMgRHU_003_416.599_425.700",
        "KTeTH31gS0k": "clip000_KTeTH31gS0k_001_766.967_829.733",
        "Kff5YkrJoBE": "clip000_Kff5YkrJoBE_006_257.933_265.167",
        "KgFKuNKL16I": "clip000_KgFKuNKL16I_001_4.004_12.246",
        "Knm0TAFACG4": "clip000_Knm0TAFACG4_002_46.800_59.267",
        "LTFbupvY06o": "clip000_LTFbupvY06o_004_97.300_105.733",
        "M6lURYXINAw": "clip000_M6lURYXINAw_004_301.067_310.633",
        "M7G0y5XC3K4": "clip000_M7G0y5XC3K4_005_242.000_249.100",
        "MKsU9tYwl2U": "clip000_MKsU9tYwl2U_001_8.100_17.940",
        "MtH54cX4enM": "clip001_MtH54cX4enM_012_925.000_989.767",
        "NJspFEtPOC8": "clip001_NJspFEtPOC8_004_348.917_576.633",
        "ODzyaMRExpA": "clip001_ODzyaMRExpA_001_176.277_253.571",
        "ON8zoj9blyw": "clip000_ON8zoj9blyw_010_1352.933_1373.817",
        "Odt5RXlH4Fc": "clip000_Odt5RXlH4Fc_004_169.867_200.833",
        "PFAyDenVpiU": "clip000_PFAyDenVpiU_005_149.167_174.733",
        "Pj3M-lAj_Jc": "clip000_Pj3M-lAj_Jc_001_130.800_184.500",
        "Pj7WX3hd-B0": "clip000_Pj7WX3hd-B0_004_71.933_98.267",
        "Pkyu6HXpBhw": "clip000_Pkyu6HXpBhw_005_153.540_221.860",
        "PsNIkztagZc": "clip000_PsNIkztagZc_001_3.333_10.433",
        "PtPZy54YCsY": "clip000_PtPZy54YCsY_002_29.033_35.900",
        "PzGW9km2KGA": "clip000_PzGW9km2KGA_006_407.567_415.133",
        "Q-R3kNAGioA": "clip000_Q-R3kNAGioA_003_105.305_117.885",
        "Q4NcwgJEO3A": "clip000_Q4NcwgJEO3A_001_474.100_483.500",
        "Q4T4xze7ltE": "clip000_Q4T4xze7ltE_008_295.733_305.867",
        "Q7IHe8lVWoE": "clip000_Q7IHe8lVWoE_001_0.100_255.656",
        "QGVfltSa2QE": "clip000_QGVfltSa2QE_001_0.100_8.567",
        "QMmE5lBjlV0": "clip000_QMmE5lBjlV0_003_41.867_48.233",
        "QQEbuZLUdcA": "clip000_QQEbuZLUdcA_002_8.100_22.100",
        "QZAWHBkBA5g": "clip000_QZAWHBkBA5g_002_41.000_57.467",
        "RNABfH_SRc4": "clip000_RNABfH_SRc4_001_25.792_43.610",
        "RX0KDjXmEkM": "clip000_RX0KDjXmEkM_003_53.700_66.383",
        "RjErLhQYBx8": "clip000_RjErLhQYBx8_002_285.051_292.042",
        "RkLo6EloqK4": "clip000_RkLo6EloqK4_002_35.233_50.233",
        "RpGbzqgQ8bA": "clip000_RpGbzqgQ8bA_001_213.667_288.000",
        "S6n2DOMjf3o": "clip001_S6n2DOMjf3o_001_248.067_752.467",
        "SBGlmG16-2Y": "clip000_SBGlmG16-2Y_001_60.817_78.833",
        "SE60vvFyVHI": "clip000_SE60vvFyVHI_015_506.580_519.980",
        "SH1i0SU6qwY": "clip000_SH1i0SU6qwY_002_89.000_210.533",
        "SQ97u06Tkd8": "clip004_SQ97u06Tkd8_002_43.433_236.400",
        "S_rPuYb88Z4": "clip000_S_rPuYb88Z4_002_53.933_76.900",
        "Se9LvevEpZU": "clip000_Se9LvevEpZU_001_0.100_34.633",
        "Ss1fcASwYT8": "clip000_Ss1fcASwYT8_001_0.100_9.359",
        "T8gm3m-7BMI": "clip000_T8gm3m-7BMI_002_21.188_34.034",
        "TF3Hb0YQwyE": "clip000_TF3Hb0YQwyE_001_0.100_393.000",
        "TJVronOA1ig": "clip014_TJVronOA1ig_002_18.533_751.567",
        "TT1z4kEkfLY": "clip000_TT1z4kEkfLY_001_0.100_55.133",
        "TYSKfE1qA3A": "clip000_TYSKfE1qA3A_002_128.633_142.267",
        "T_xrjQIh39E": "clip000_T_xrjQIh39E_008_149.167_169.633",
        "UAw7Eq4PCyo": "clip000_UAw7Eq4PCyo_004_176.176_214.348",
        "UBZXTSpMkfQ": "clip001_UBZXTSpMkfQ_044_1245.733_1312.800",
        "UkDQi3BTOtc": "clip001_UkDQi3BTOtc_002_41.867_79.700",
        "WFN70lLng20": "clip000_WFN70lLng20_001_0.100_21.221",
        "Y8zklqlpBT8": "clip001_Y8zklqlpBT8_004_54.367_100.933",
        "YAsBMTRmgK4": "clip000_YAsBMTRmgK4_012_442.272_456.107",
        "YJquBPoczZY": "clip000_YJquBPoczZY_011_323.723_342.275",
        "YVm8gySFY6c": "clip000_YVm8gySFY6c_001_23.400_30.033",
        "Ykx6gaEyTGs": "clip000_Ykx6gaEyTGs_001_149.567_160.533",
        "Z0jJIB7iU2A": "clip000_Z0jJIB7iU2A_010_154.100_160.580",
        "Z1wAbRH3Pr4": "clip000_Z1wAbRH3Pr4_001_701.500_1013.200",
        "ZJYpTM6ol7g": "clip007_ZJYpTM6ol7g_006_333.267_595.300",
        "Zd_0e5Pzjbg": "clip000_Zd_0e5Pzjbg_002_72.773_81.215",
        "ZfNLlFTEvXs": "clip001_ZfNLlFTEvXs_002_222.072_265.374",
        "ZkCeywIAMbU": "clip001_ZkCeywIAMbU_001_9.933_59.967",
        "Zo0n-lCjTNg": "clip000_Zo0n-lCjTNg_003_84.284_153.003",
        "_AWO54kIMVI": "clip000__AWO54kIMVI_003_217.767_225.509",
        "_C3Kq-2R-ZU": "clip000__C3Kq-2R-ZU_001_6.808_42.067",
        "_R5F2XyPhoo": "clip002__R5F2XyPhoo_006_89.033_343.967",
        "a_AGPRTGBJ4": "clip000_a_AGPRTGBJ4_006_249.833_275.433",
        "ak8d3ZZ_LA0": "clip000_ak8d3ZZ_LA0_013_817.033_833.567",
        "at9AapgKkdM": "clip000_at9AapgKkdM_003_234.820_243.100",
        "azDWX0qBzBU": "clip000_azDWX0qBzBU_001_123.967_131.800",
        "b54CvizC2RQ": "clip001_b54CvizC2RQ_001_102.040_139.840",
        "bDYPHRyFa9M": "clip000_bDYPHRyFa9M_003_36.460_44.060",
        "bQohY8RRDEM": "clip000_bQohY8RRDEM_002_61.633_153.900",
        "bScwkAeT3PU": "clip001_bScwkAeT3PU_001_0.100_344.667",
        "b_60DZnl0iU": "clip000_b_60DZnl0iU_006_1039.140_1054.220",
        "bauYD4rgv9o": "clip000_bauYD4rgv9o_004_58.725_87.888",
        "bsGAqKZLTLE": "clip000_bsGAqKZLTLE_001_3.300_21.600",
        "bxxN3KkM66U": "clip000_bxxN3KkM66U_001_23.090_42.226",
        "c-X8DJye6Gw": "clip000_c-X8DJye6Gw_002_202.400_220.033",
        "c1tA412HLbk": "clip000_c1tA412HLbk_003_90.433_208.600",
        "c7luHlz0j0g": "clip000_c7luHlz0j0g_006_128.361_144.077",
        "cO18Qe8ymsQ": "clip000_cO18Qe8ymsQ_003_55.906_88.197",
        "cpRfY_-myyU": "clip003_cpRfY_-myyU_002_409.600_602.333",
        "cxP1xBUToWs": "clip000_cxP1xBUToWs_001_27.044_33.475",
        "czdLAdtE_74": "clip000_czdLAdtE_74_003_134.200_141.133",
        "d-X9UiCCrYk": "clip000_d-X9UiCCrYk_001_42.700_65.800",
        "d23PCP_ZLmY": "clip000_d23PCP_ZLmY_004_64.960_74.340",
        "d6kl1UwtroQ": "clip000_d6kl1UwtroQ_001_0.100_57.300",
        "dgBb4C-_zgM": "clip005_dgBb4C-_zgM_001_247.142_547.900",
        "e0z2fc2tF48": "clip000_e0z2fc2tF48_001_6.433_75.317",
        "eAKh9rNTc-8": "clip000_eAKh9rNTc-8_002_51.167_60.400",
        "eIL64MLLhx8": "clip000_eIL64MLLhx8_001_0.100_9.833",
        "eMRgXi5vTOY": "clip000_eMRgXi5vTOY_002_34.433_50.942",
        "ePXUK9NlPr0": "clip000_ePXUK9NlPr0_001_7.867_18.600",
        "e_UifoTC31g": "clip000_e_UifoTC31g_001_0.100_22.900",
        "eu8qZAPcacQ": "clip000_eu8qZAPcacQ_001_0.100_6.700",
        "exs8bpwyDnY": "clip000_exs8bpwyDnY_002_5.933_27.933",
        "fKbpJaVT2X8": "clip000_fKbpJaVT2X8_004_48.300_55.167",
        "ftEz2dnAR-A": "clip000_ftEz2dnAR-A_001_55.500_75.033",
        "fwtA1MmbSlE": "clip002_fwtA1MmbSlE_001_859.933_962.000",
        "fyURLxTKUe4": "clip000_fyURLxTKUe4_002_453.860_501.540",
        "g0RWoZnOANM": "clip000_g0RWoZnOANM_001_0.100_9.967",
        "g1oydeZdHS8": "clip000_g1oydeZdHS8_006_375.983_387.583",
        "gPldXYi2WWs": "clip000_gPldXYi2WWs_004_232.200_279.600",
        "gUN207IqziQ": "clip000_gUN207IqziQ_005_109.540_116.500",
        "g_cXR9HgUL4": "clip000_g_cXR9HgUL4_001_0.100_18.000",
        "gi_FY6GL6yo": "clip000_gi_FY6GL6yo_001_0.100_37.900",
        "gymo4hRZeO4": "clip000_gymo4hRZeO4_003_118.540_126.220",
        "h2Jgu5-DY2s": "clip000_h2Jgu5-DY2s_002_20.850_42.108",
        "h375KdmrA6Y": "clip000_h375KdmrA6Y_012_331.798_343.276",
        "hAL5VjfpAyU": "clip000_hAL5VjfpAyU_003_153.033_169.000",
        "hVpDFmcul84": "clip000_hVpDFmcul84_010_567.967_628.100",
        "hc_OVXyNic0": "clip000_hc_OVXyNic0_001_22.767_33.867",
        "heqkYBDUUB8": "clip000_heqkYBDUUB8_001_0.100_18.150",
        "hsYMLPff95U": "clip000_hsYMLPff95U_002_309.233_343.600",
        "iJoH68s_4hU": "clip000_iJoH68s_4hU_006_141.433_163.200",
        "jCwTSgDkM9o": "clip000_jCwTSgDkM9o_002_118.583_133.700",
        "jIKA6hqxj6o": "clip001_jIKA6hqxj6o_001_0.100_399.500",
        "jOAaVpFKCHM": "clip000_jOAaVpFKCHM_004_428.795_450.951",
        "j_ov5aKgDTI": "clip001_j_ov5aKgDTI_002_60.327_109.776",
        "kxaSt5JfoCA": "clip000_kxaSt5JfoCA_002_16.100_24.033",
        "lXGECvyAjc0": "clip000_lXGECvyAjc0_005_242.700_249.367",
        "nYualbpDRis": "clip000_nYualbpDRis_003_281.147_303.203",
        "oWo2ej_jbvA": "clip000_oWo2ej_jbvA_001_276.100_287.900",
        "oqSEMKGHlX4": "clip000_oqSEMKGHlX4_002_199.132_216.116",
        "pPwuBjT5hzY": "clip000_pPwuBjT5hzY_002_37.517_79.650",
        "pYW4fStQxSc": "clip000_pYW4fStQxSc_005_41.167_58.133",
        "pcJV7A1IVw4": "clip000_pcJV7A1IVw4_001_17.067_27.233",
        "pnQCfACUeSo": "clip000_pnQCfACUeSo_003_547.233_555.667",
        "psUAg3VsOBk": "clip000_psUAg3VsOBk_001_0.100_85.533",
        "pvPJQ8dSCEo": "clip000_pvPJQ8dSCEo_004_65.000_73.033",
        "pxWhygMq6KQ": "clip000_pxWhygMq6KQ_001_0.100_15.420",
        "pxju2jy479I": "clip002_pxju2jy479I_005_202.300_272.400",
        "pyDJVXVzmbY": "clip000_pyDJVXVzmbY_001_7.380_16.760",
        "q0HOjVAjWDI": "clip001_q0HOjVAjWDI_001_32.967_87.500",
        "qEWb6oRA260": "clip000_qEWb6oRA260_002_10.644_18.285",
        "qTp8FTM6uds": "clip000_qTp8FTM6uds_013_335.867_355.833",
        "r0pxv8LTMg4": "clip001_r0pxv8LTMg4_002_138.505_175.142",
        "sBbE2Gd6syc": "clip000_sBbE2Gd6syc_014_570.420_582.180",
        "vSqoWFauosM": "clip000_vSqoWFauosM_002_367.334_380.580",
        "w9n1lWyyzc4": "clip000_w9n1lWyyzc4_047_884.333_897.167"
      },
      "top_sources_by_clips": [
        [
          "g0RWoZnOANM",
          35
        ],
        [
          "9CRI6CMkMok",
          27
        ],
        [
          "pyDJVXVzmbY",
          24
        ],
        [
          "TF3Hb0YQwyE",
          22
        ],
        [
          "9eOGgAKXm4Y",
          20
        ],
        [
          "Se9LvevEpZU",
          20
        ],
        [
          "ak8d3ZZ_LA0",
          20
        ],
        [
          "bsGAqKZLTLE",
          19
        ],
        [
          "4R1yqW6RFck",
          18
        ],
        [
          "6NKO4foW8oQ",
          15
        ]
      ]
    },
    "JoyGen": {
      "name": "JoyGen",
      "split_basis": "train.txt",
      "train_list_count": 1245,
      "meta_json_count": 1506,
      "inventory_clip_count": 1506,
      "inventory_meta_without_mp4_count": 0,
      "usable_clip_count": 1245,
      "train_missing_mp4_count": 0,
      "train_missing_meta_count": 0,
      "total_duration_seconds": 12293.12,
      "total_duration_human": "3.41 h",
      "duration_stats_seconds": {
        "mean": 9.874,
        "median": 10.0,
        "min": 6.0,
        "max": 19.76
      },
      "unique_speakers": 91,
      "clips_per_speaker": {
        "mean": 13.68,
        "median": 8,
        "max": 80
      },
      "speaker_count_method": "Grouped by the source identifier before _shot_.",
      "resolution": [
        "1080x1440",
        "1080x1904",
        "1080x1906",
        "1080x1920",
        "1440x1072",
        "1440x1080",
        "1888x1062",
        "1920x1078",
        "1920x1080",
        "720x1280",
        "886x1920"
      ],
      "fps": [
        25.0
      ],
      "frame_count_stats": {
        "mean": 246.85,
        "median": 250,
        "min": 150,
        "max": 494
      },
      "face_box_stats": {
        "mean_width": 410.0,
        "mean_height": 524.19,
        "median_width": 383,
        "median_height": 513,
        "mean_area_ratio_percent": 11.12,
        "median_area_ratio_percent": 9.75
      },
      "modalities": {
        "train_mp4_clips": 1245,
        "train_wav_clips": 1245,
        "train_wav_16k_clips": 0,
        "inventory_wav_clips": 1506,
        "inventory_wav_16k_clips": 0,
        "original_wav_sample_rates": [
          16000
        ],
        "original_wav_channels": [
          1
        ],
        "original_wav_bits": [
          16
        ],
        "wav_16k_sample_rates": []
      },
      "metadata_schema": [
        "face_list",
        "face_size",
        "frames",
        "isvalid",
        "landmark_list",
        "mp4_path",
        "video_size",
        "wav_path"
      ],
      "format_description": {
        "root": "dataset/JoyGen",
        "meta_dir": "dataset/JoyGen/meta",
        "media_dir": "dataset/JoyGen/video_audio_clip_root",
        "split_file": "dataset/JoyGen/train.txt",
        "pairing_rule": "One JSON metadata file and one MP4 clip share the same stem; each clip also has paired WAV audio.",
        "json_fields": [
          "face_list",
          "face_size",
          "frames",
          "isvalid",
          "landmark_list",
          "mp4_path",
          "video_size",
          "wav_path"
        ],
        "json_video_size_order": "[height, width]",
        "json_face_size_order": "[height, width]",
        "annotations": [
          "per-frame face bounding boxes",
          "per-frame facial landmarks",
          "frame count",
          "paired audio/video paths",
          "face crop size",
          "validity flag"
        ]
      },
      "speaker_examples": {
        "Bili_0001": "Bili_0001_shot_001_001",
        "Bili_0003": "Bili_0003_shot_001_000",
        "Bili_0004": "Bili_0004_shot_001_000",
        "Bili_0006": "Bili_0006_shot_001_004",
        "Bili_0014": "Bili_0014_shot_012_002",
        "Bili_0018": "Bili_0018_shot_002_000",
        "Bili_0032": "Bili_0032_shot_006_000",
        "Bili_0036": "Bili_0036_shot_007_000",
        "Bili_0037": "Bili_0037_shot_091_000",
        "Bili_0040": "Bili_0040_shot_001_000",
        "Bili_0043": "Bili_0043_shot_003_000",
        "Bili_0046": "Bili_0046_shot_001_000",
        "Bili_0047": "Bili_0047_shot_001_000",
        "Bili_0052": "Bili_0052_shot_003_000",
        "Bili_0056": "Bili_0056_shot_001_000",
        "Bili_0063": "Bili_0063_shot_025_000",
        "Bili_0077": "Bili_0077_shot_002_001",
        "Bili_0081": "Bili_0081_shot_052_005",
        "Bili_0087": "Bili_0087_shot_005_007",
        "Bili_0094": "Bili_0094_shot_001_000",
        "Bili_0095": "Bili_0095_shot_001_021",
        "Bili_0098": "Bili_0098_shot_001_003",
        "Bili_0103": "Bili_0103_shot_002_000",
        "Bili_0119": "Bili_0119_shot_002_001",
        "Bili_0123": "Bili_0123_shot_007_001",
        "Bili_0124": "Bili_0124_shot_001_006",
        "Bili_0127": "Bili_0127_shot_001_000",
        "Bili_0129": "Bili_0129_shot_004_000",
        "Bili_0133": "Bili_0133_shot_002_000",
        "Bili_0192": "Bili_0192_shot_001_000",
        "Bili_0225": "Bili_0225_shot_003_000",
        "Bili_0263": "Bili_0263_shot_003_030",
        "Bili_0265": "Bili_0265_shot_001_000",
        "Bili_0267": "Bili_0267_shot_003_000",
        "Bili_0273": "Bili_0273_shot_018_001",
        "Bili_0284": "Bili_0284_shot_001_005",
        "Bili_0286": "Bili_0286_shot_002_002",
        "Bili_0292": "Bili_0292_shot_021_000",
        "Bili_0299": "Bili_0299_shot_001_001",
        "Bili_0305": "Bili_0305_shot_002_001",
        "Bili_0320": "Bili_0320_shot_001_000",
        "Bili_0324": "Bili_0324_shot_003_000",
        "Bili_0332": "Bili_0332_shot_004_000",
        "Bili_0335": "Bili_0335_shot_003_008",
        "Bili_0340": "Bili_0340_shot_001_000",
        "Bili_0358": "Bili_0358_shot_004_001",
        "Bili_0359": "Bili_0359_shot_003_000",
        "Bili_0361": "Bili_0361_shot_001_000",
        "Bili_0363": "Bili_0363_shot_001_000",
        "Bili_0367": "Bili_0367_shot_005_005",
        "Bili_0368": "Bili_0368_shot_001_001",
        "Bili_0387": "Bili_0387_shot_002_012",
        "Bili_0400": "Bili_0400_shot_001_001",
        "Bili_0410": "Bili_0410_shot_001_001",
        "Bili_0412": "Bili_0412_shot_001_008",
        "Bili_0439": "Bili_0439_shot_002_003",
        "Bili_0495": "Bili_0495_shot_001_006",
        "Bili_0509": "Bili_0509_shot_002_000",
        "Bili_0521": "Bili_0521_shot_001_035",
        "Bili_0540": "Bili_0540_shot_003_000",
        "Bili_0542": "Bili_0542_shot_005_001",
        "Bili_0557": "Bili_0557_shot_001_001",
        "Bili_0558": "Bili_0558_shot_004_012",
        "Bili_0560": "Bili_0560_shot_001_001",
        "Bili_0562": "Bili_0562_shot_001_002",
        "Bili_0567": "Bili_0567_shot_002_003",
        "Bili_0576": "Bili_0576_shot_065_000",
        "Bili_0580": "Bili_0580_shot_005_002",
        "Bili_0582": "Bili_0582_shot_002_001",
        "Bili_0583": "Bili_0583_shot_001_000",
        "Bili_0591": "Bili_0591_shot_005_000",
        "Bili_0598": "Bili_0598_shot_011_000",
        "Bili_0618": "Bili_0618_shot_001_032",
        "Bili_0621": "Bili_0621_shot_001_000",
        "Bili_0625": "Bili_0625_shot_001_003",
        "Bili_0634": "Bili_0634_shot_001_000",
        "Bili_0644": "Bili_0644_shot_001_001",
        "Bili_0648": "Bili_0648_shot_002_000",
        "Bili_0651": "Bili_0651_shot_003_000",
        "Bili_0659": "Bili_0659_shot_011_000",
        "Bili_0670": "Bili_0670_shot_003_002",
        "SaveTik.io_7336869647305887027_hd": "SaveTik.io_7336869647305887027_hd_shot_002_000",
        "SaveTik.io_7342776901041196297_hd": "SaveTik.io_7342776901041196297_hd_shot_001_000",
        "SaveTik.io_7344678214398151946_hd": "SaveTik.io_7344678214398151946_hd_shot_003_000",
        "SaveTik.io_7350666335048715572_hd": "SaveTik.io_7350666335048715572_hd_shot_001_002",
        "SaveTik.io_7358030605755469067_hd": "SaveTik.io_7358030605755469067_hd_shot_001_001",
        "SaveTik.io_7365124591385169189_hd": "SaveTik.io_7365124591385169189_hd_shot_001_000",
        "SaveTik.io_7365821285953244466_hd": "SaveTik.io_7365821285953244466_hd_shot_001_000",
        "SaveTik.io_7366538093240896820_hd": "SaveTik.io_7366538093240896820_hd_shot_003_000",
        "SaveTik.io_7366969439717297459_hd": "SaveTik.io_7366969439717297459_hd_shot_002_000",
        "SaveTik.io_7369080544820268307_hd": "SaveTik.io_7369080544820268307_hd_shot_001_001"
      },
      "top_sources_by_clips": [
        [
          "Bili_0670",
          80
        ],
        [
          "Bili_0077",
          66
        ],
        [
          "Bili_0046",
          55
        ],
        [
          "Bili_0103",
          54
        ],
        [
          "Bili_0410",
          48
        ],
        [
          "SaveTik.io_7366969439717297459_hd",
          43
        ],
        [
          "Bili_0583",
          37
        ],
        [
          "Bili_0299",
          35
        ],
        [
          "SaveTik.io_7365124591385169189_hd",
          34
        ],
        [
          "Bili_0098",
          29
        ]
      ]
    }
  },
  "combined": {
    "usable_clip_count": 2262,
    "unique_speakers": 325,
    "total_duration_seconds": 32336.08,
    "total_duration_human": "8.98 h",
    "datasets": [
      "TalkVid",
      "JoyGen"
    ]
  }
}