{
  "date": "2026-11-13",
  "source": "Whisper Summary",
  "results": [
    {
      "config": "C3 CPU acceptance nemo_sortformer",
      "duration": "3.0s",
      "asr_time,rtf": "0.4s,0.1263",
      "diarization_time,rtf": "0.2s,0.0592",
      "total_rtf": "0.2245",
      "artifact": "wav / 0.1 MB",
      "speakers": "1",
      "turns": "1",
      "switches": "0",
      "unknown": "0"
    },
    {
      "config": "C3 CPU acceptance sherpa_onnx",
      "duration": "3.0s",
      "asr_time,rtf": "0.4s,0.1236",
      "diarization_time,rtf": "0.2s,0.0549",
      "total_rtf": "0.2149",
      "artifact": "wav / 0.1 MB",
      "speakers": "1",
      "turns": "1",
      "switches": "0",
      "unknown": "0"
    },
    {
      "config": "C3 CPU acceptance pyannote",
      "duration": "3.0s",
      "asr_time,rtf": "0.4s,0.1279",
      "diarization_time,rtf": "0.2s,0.0734",
      "total_rtf": "0.2843",
      "artifact": "wav / 0.1 MB",
      "speakers": "1",
      "turns": "1",
      "switches": "0",
      "unknown": "0"
    },
    {
      "config": "Sortformer very-high-latency batch 1 Russian check",
      "duration": "3.0s",
      "asr_time,rtf": "0.4s,0.1319",
      "diarization_time,rtf": "0.2s,0.0552",
      "total_rtf": "0.3094",
      "artifact": "aac_transcode / 0.0 MB",
      "speakers": "1",
      "turns": "1",
      "switches": "0",
      "unknown": "0"
    },
    {
      "config": "sherpa-onnx default Russian check",
      "duration": "3.0s",
      "asr_time,rtf": "0.4s,0.1293",
      "diarization_time,rtf": "0.1s,0.0489",
      "total_rtf": "0.2899",
      "artifact": "aac_transcode / 0.0 MB",
      "speakers": "1",
      "turns": "1",
      "switches": "0",
      "unknown": "0"
    },
    {
      "config": "pyannote isolated batch 2",
      "duration": "3.0s",
      "asr_time,rtf": "0.4s,0.1336",
      "diarization_time,rtf": "0.2s,0.0787",
      "total_rtf": "0.3369",
      "artifact": "aac_transcode / 0.0 MB",
      "speakers": "1",
      "turns": "1",
      "switches": "0",
      "unknown": "0"
    },
    {
      "config": "pyannote selected shared batch 2",
      "duration": "3.0s",
      "asr_time,rtf": "0.4s,0.1398",
      "diarization_time,rtf": "0.2s,0.0714",
      "total_rtf": "0.4374",
      "artifact": "aac_transcode / 0.0 MB",
      "speakers": "1",
      "turns": "1",
      "switches": "0",
      "unknown": "0"
    },
    {
      "config": "Sortformer very-high-latency batch 1 Russian check",
      "duration": "9.3s",
      "asr_time,rtf": "0.5s,0.0556",
      "diarization_time,rtf": "0.2s,0.0182",
      "total_rtf": "0.2314",
      "artifact": "aac_transcode / 0.1 MB",
      "speakers": "2",
      "turns": "3",
      "switches": "2",
      "unknown": "0"
    },
    {
      "config": "sherpa-onnx default Russian check",
      "duration": "9.3s",
      "asr_time,rtf": "0.5s,0.0532",
      "diarization_time,rtf": "0.2s,0.0170",
      "total_rtf": "0.1810",
      "artifact": "aac_transcode / 0.1 MB",
      "speakers": "2",
      "turns": "3",
      "switches": "2",
      "unknown": "0"
    },
    {
      "config": "pyannote isolated batch 2",
      "duration": "9.3s",
      "asr_time,rtf": "0.6s,0.0593",
      "diarization_time,rtf": "0.2s,0.0259",
      "total_rtf": "0.2451",
      "artifact": "aac_transcode / 0.1 MB",
      "speakers": "2",
      "turns": "3",
      "switches": "2",
      "unknown": "0"
    },
    {
      "config": "pyannote selected shared batch 2",
      "duration": "9.3s",
      "asr_time,rtf": "0.6s,0.0665",
      "diarization_time,rtf": "0.2s,0.0237",
      "total_rtf": "0.2156",
      "artifact": "aac_transcode / 0.1 MB",
      "speakers": "2",
      "turns": "3",
      "switches": "2",
      "unknown": "0"
    },
    {
      "config": "CPU standby acceptance nemo_sortformer",
      "duration": "9.3s",
      "asr_time,rtf": "0.6s,0.0616",
      "diarization_time,rtf": "15.8s,1.6910",
      "total_rtf": "1.7663",
      "artifact": "wav / 0.3 MB",
      "speakers": "2",
      "turns": "3",
      "switches": "2",
      "unknown": "0"
    },
    {
      "config": "CPU standby acceptance sherpa_onnx",
      "duration": "9.3s",
      "asr_time,rtf": "0.5s,0.0525",
      "diarization_time,rtf": "0.3s,0.0325",
      "total_rtf": "0.0977",
      "artifact": "wav / 0.3 MB",
      "speakers": "2",
      "turns": "3",
      "switches": "2",
      "unknown": "0"
    },
    {
      "config": "CPU standby acceptance pyannote",
      "duration": "9.3s",
      "asr_time,rtf": "0.6s,0.0613",
      "diarization_time,rtf": "54.3s,5.8215",
      "total_rtf": "5.9220",
      "artifact": "wav / 0.3 MB",
      "speakers": "2",
      "turns": "3",
      "switches": "2",
      "unknown": "0"
    },
    {
      "config": "CPU standby acceptance nemo_sortformer",
      "duration": "9.3s",
      "asr_time,rtf": "0.6s,0.0592",
      "diarization_time,rtf": "14.0s,1.4999",
      "total_rtf": "1.5881",
      "artifact": "wav / 0.3 MB",
      "speakers": "2",
      "turns": "3",
      "switches": "2",
      "unknown": "0"
    },
    {
      "config": "CPU standby acceptance sherpa_onnx",
      "duration": "9.3s",
      "asr_time,rtf": "0.5s,0.0577",
      "diarization_time,rtf": "0.3s,0.0325",
      "total_rtf": "0.1207",
      "artifact": "wav / 0.3 MB",
      "speakers": "2",
      "turns": "3",
      "switches": "2",
      "unknown": "0"
    },
    {
      "config": "CPU standby acceptance pyannote",
      "duration": "9.3s",
      "asr_time,rtf": "0.5s,0.0545",
      "diarization_time,rtf": "54.0s,5.7964",
      "total_rtf": "5.8811",
      "artifact": "wav / 0.3 MB",
      "speakers": "2",
      "turns": "3",
      "switches": "2",
      "unknown": "0"
    },
    {
      "config": "Sortformer very-high-latency batch 1 Russian check",
      "duration": "9.3s",
      "asr_time,rtf": "0.5s,0.0545",
      "diarization_time,rtf": "0.2s,0.0172",
      "total_rtf": "0.1437",
      "artifact": "aac_transcode / 0.1 MB",
      "speakers": "2",
      "turns": "3",
      "switches": "2",
      "unknown": "0"
    },
    {
      "config": "sherpa-onnx default Russian check",
      "duration": "9.3s",
      "asr_time,rtf": "0.6s,0.0605",
      "diarization_time,rtf": "0.2s,0.0178",
      "total_rtf": "0.1548",
      "artifact": "aac_transcode / 0.1 MB",
      "speakers": "2",
      "turns": "3",
      "switches": "2",
      "unknown": "0"
    },
    {
      "config": "pyannote isolated batch 2",
      "duration": "9.3s",
      "asr_time,rtf": "0.5s,0.0554",
      "diarization_time,rtf": "0.3s,0.0268",
      "total_rtf": "0.1587",
      "artifact": "aac_transcode / 0.1 MB",
      "speakers": "2",
      "turns": "3",
      "switches": "2",
      "unknown": "0"
    },
    {
      "config": "pyannote selected shared batch 2",
      "duration": "9.3s",
      "asr_time,rtf": "0.6s,0.0629",
      "diarization_time,rtf": "0.2s,0.0208",
      "total_rtf": "0.1670",
      "artifact": "aac_transcode / 0.1 MB",
      "speakers": "2",
      "turns": "3",
      "switches": "2",
      "unknown": "0"
    },
    {
      "config": "CPU standby acceptance sherpa_onnx",
      "duration": "9.3s",
      "asr_time,rtf": "0.5s,0.0585",
      "diarization_time,rtf": "0.3s,0.0323",
      "total_rtf": "0.1129",
      "artifact": "wav / 0.3 MB",
      "speakers": "2",
      "turns": "3",
      "switches": "2",
      "unknown": "0"
    },
    {
      "config": "CPU standby acceptance pyannote",
      "duration": "9.3s",
      "asr_time,rtf": "0.5s,0.0587",
      "diarization_time,rtf": "1.2m,7.5662",
      "total_rtf": "7.6325",
      "artifact": "wav / 0.3 MB",
      "speakers": "2",
      "turns": "3",
      "switches": "2",
      "unknown": "0"
    },
    {
      "config": "CPU standby acceptance nemo_sortformer",
      "duration": "9.3s",
      "asr_time,rtf": "0.6s,0.0590",
      "diarization_time,rtf": "42.3s,4.5342",
      "total_rtf": "4.6042",
      "artifact": "wav / 0.3 MB",
      "speakers": "2",
      "turns": "3",
      "switches": "2",
      "unknown": "0"
    },
    {
      "config": "Sortformer very-high-latency batch 1 Russian check",
      "duration": "9.3s",
      "asr_time,rtf": "0.6s,0.0613",
      "diarization_time,rtf": "0.2s,0.0182",
      "total_rtf": "0.1152",
      "artifact": "aac_transcode / 0.1 MB",
      "speakers": "2",
      "turns": "3",
      "switches": "2",
      "unknown": "0"
    },
    {
      "config": "sherpa-onnx default Russian check",
      "duration": "9.3s",
      "asr_time,rtf": "0.6s,0.0599",
      "diarization_time,rtf": "0.1s,0.0151",
      "total_rtf": "0.1090",
      "artifact": "aac_transcode / 0.1 MB",
      "speakers": "2",
      "turns": "3",
      "switches": "2",
      "unknown": "0"
    },
    {
      "config": "pyannote isolated batch 2",
      "duration": "9.3s",
      "asr_time,rtf": "0.6s,0.0598",
      "diarization_time,rtf": "0.2s,0.0256",
      "total_rtf": "0.1204",
      "artifact": "aac_transcode / 0.1 MB",
      "speakers": "2",
      "turns": "3",
      "switches": "2",
      "unknown": "0"
    },
    {
      "config": "pyannote selected shared batch 2",
      "duration": "9.3s",
      "asr_time,rtf": "0.6s,0.0622",
      "diarization_time,rtf": "0.2s,0.0208",
      "total_rtf": "0.1258",
      "artifact": "aac_transcode / 0.1 MB",
      "speakers": "2",
      "turns": "3",
      "switches": "2",
      "unknown": "0"
    },
    {
      "config": "pyannote shared batch 2 auto",
      "duration": "18.5m",
      "asr_time,rtf": "37.4s,0.0337",
      "diarization_time,rtf": "45.2s,0.0407",
      "total_rtf": "0.0895",
      "artifact": "aac_transcode / 9.2 MB",
      "speakers": "3",
      "turns": "101",
      "switches": "80",
      "unknown": "3"
    },
    {
      "config": "pyannote shared batch 2 auto",
      "duration": "18.5m",
      "asr_time,rtf": "39.9s,0.0360",
      "diarization_time,rtf": "45.1s,0.0407",
      "total_rtf": "0.0914",
      "artifact": "aac_transcode / 9.2 MB",
      "speakers": "3",
      "turns": "98",
      "switches": "76",
      "unknown": "2"
    },
    {
      "config": "pyannote shared batch 2 auto",
      "duration": "18.5m",
      "asr_time,rtf": "39.3s,0.0355",
      "diarization_time,rtf": "43.0s,0.0388",
      "total_rtf": "0.0884",
      "artifact": "aac_transcode / 9.2 MB",
      "speakers": "3",
      "turns": "101",
      "switches": "80",
      "unknown": "3"
    },
    {
      "config": "pyannote shared batch 2 auto",
      "duration": "58.0m",
      "asr_time,rtf": "1.7m,0.0297",
      "diarization_time,rtf": "2.4m,0.0411",
      "total_rtf": "0.0827",
      "artifact": "aac_transcode / 28.2 MB",
      "speakers": "2",
      "turns": "227",
      "switches": "173",
      "unknown": "7"
    },
    {
      "config": "Sortformer very high_latency batch 1",
      "duration": "9.6m",
      "asr_time,rtf": "16.7s,0.0291",
      "diarization_time,rtf": "5.4s,0.0094",
      "total_rtf": "0.0457",
      "artifact": "source_aac_copy / 8.9 MB",
      "speakers": "4",
      "turns": "121",
      "switches": "99",
      "unknown": "11"
    },
    {
      "config": "sherpa-onnx threshold 0.5",
      "duration": "9.6m",
      "asr_time,rtf": "17.1s,0.0298",
      "diarization_time,rtf": "1.3m,0.1375",
      "total_rtf": "0.1742",
      "artifact": "source_aac_copy / 8.9 MB",
      "speakers": "6",
      "turns": "79",
      "switches": "52",
      "unknown": "4"
    },
    {
      "config": "pyannote shared batch 2 auto",
      "duration": "9.6m",
      "asr_time,rtf": "18.9s,0.0328",
      "diarization_time,rtf": "24.8s,0.0432",
      "total_rtf": "0.0838",
      "artifact": "source_aac_copy / 8.9 MB",
      "speakers": "7",
      "turns": "106",
      "switches": "90",
      "unknown": "2"
    },
    {
      "config": "final Sortformer chunked v10 big",
      "duration": "1.79h",
      "asr_time,rtf": "2.3m,0.0215",
      "diarization_time,rtf": "25.5s,0.0039",
      "total_rtf": "0.0472",
      "artifact": "aac_transcode / 47.0 MB",
      "speakers": "4",
      "turns": "618",
      "switches": "303",
      "unknown": "93"
    },
    {
      "config": "final pyannote shared batch 2 auto big",
      "duration": "1.79h",
      "asr_time,rtf": "2.5m,0.0235",
      "diarization_time,rtf": "4.3m,0.0400",
      "total_rtf": "0.0897",
      "artifact": "aac_transcode / 47.0 MB",
      "speakers": "8",
      "turns": "614",
      "switches": "298",
      "unknown": "24"
    },
    {
      "config": "final Sortformer very-high-latency batch 1",
      "duration": "9.6m",
      "asr_time,rtf": "18.3s,0.0318",
      "diarization_time,rtf": "4.7s,0.0082",
      "total_rtf": "0.0476",
      "artifact": "source_aac_copy / 8.9 MB",
      "speakers": "4",
      "turns": "121",
      "switches": "99",
      "unknown": "11"
    },
    {
      "config": "final sherpa-onnx clusters 7",
      "duration": "9.6m",
      "asr_time,rtf": "20.0s,0.0348",
      "diarization_time,rtf": "1.3m,0.1395",
      "total_rtf": "0.1818",
      "artifact": "source_aac_copy / 8.9 MB",
      "speakers": "6",
      "turns": "79",
      "switches": "52",
      "unknown": "4"
    },
    {
      "config": "final pyannote shared batch 2 auto",
      "duration": "9.6m",
      "asr_time,rtf": "20.2s,0.0351",
      "diarization_time,rtf": "27.1s,0.0471",
      "total_rtf": "0.0901",
      "artifact": "source_aac_copy / 8.9 MB",
      "speakers": "7",
      "turns": "106",
      "switches": "90",
      "unknown": "2"
    },
    {
      "config": "sherpa-onnx default Russian check",
      "duration": "9.6m",
      "asr_time,rtf": "15.9s,0.0277",
      "diarization_time,rtf": "1.4m,0.1450",
      "total_rtf": "0.1805",
      "artifact": "source_aac_copy / 8.9 MB",
      "speakers": "58",
      "turns": "95",
      "switches": "83",
      "unknown": "5"
    },
    {
      "config": "pyannote selected shared batch 2",
      "duration": "9.6m",
      "asr_time,rtf": "18.3s,0.0318",
      "diarization_time,rtf": "28.2s,0.0490",
      "total_rtf": "0.0895",
      "artifact": "source_aac_copy / 8.9 MB",
      "speakers": "7",
      "turns": "106",
      "switches": "90",
      "unknown": "2"
    },
    {
      "config": "pyannote AAC transcode shared MIG",
      "duration": "18.5m",
      "asr_time,rtf": "38.9s,0.0351",
      "diarization_time,rtf": "56.4s,0.0509",
      "total_rtf": "0.0996",
      "artifact": "aac_transcode / 9.2 MB",
      "speakers": "3",
      "turns": "102",
      "switches": "81",
      "unknown": "4"
    },
    {
      "config": "pyannote isolated MIG batch 8",
      "duration": "1.79h",
      "asr_time,rtf": "4.4m,0.0405",
      "diarization_time,rtf": "3.6m,0.0338",
      "total_rtf": "0.0942",
      "artifact": "aac_transcode / 47.0 MB",
      "speakers": "8",
      "turns": "614",
      "switches": "298",
      "unknown": "24"
    },
    {
      "config": "pyannote isolated MIG batch 2",
      "duration": "1.79h",
      "asr_time,rtf": "3.8m,0.0351",
      "diarization_time,rtf": "4.7m,0.0438",
      "total_rtf": "0.0984",
      "artifact": "aac_transcode / 47.0 MB",
      "speakers": "8",
      "turns": "614",
      "switches": "298",
      "unknown": "24"
    },
    {
      "config": "pyannote isolated MIG batch 8",
      "duration": "9.6m",
      "asr_time,rtf": "34.3s,0.0596",
      "diarization_time,rtf": "23.2s,0.0404",
      "total_rtf": "0.1073",
      "artifact": "source_aac_copy / 8.9 MB",
      "speakers": "7",
      "turns": "106",
      "switches": "90",
      "unknown": "2"
    },
    {
      "config": "pyannote isolated MIG batch 4",
      "duration": "9.6m",
      "asr_time,rtf": "43.7s,0.0759",
      "diarization_time,rtf": "5.6m,0.5837",
      "total_rtf": "0.6663",
      "artifact": "source_aac_copy / 8.9 MB",
      "speakers": "7",
      "turns": "106",
      "switches": "90",
      "unknown": "2"
    },
    {
      "config": "pyannote isolated MIG batch 2",
      "duration": "9.6m",
      "asr_time,rtf": "23.9s,0.0415",
      "diarization_time,rtf": "25.9s,0.0451",
      "total_rtf": "0.0941",
      "artifact": "source_aac_copy / 8.9 MB",
      "speakers": "7",
      "turns": "106",
      "switches": "90",
      "unknown": "2"
    },
    {
      "config": "pyannote auto audio hints 2-8 shared MIG",
      "duration": "9.6m",
      "asr_time,rtf": "19.1s,0.0332",
      "diarization_time,rtf": "46.6s,0.0810",
      "total_rtf": "0.1213",
      "artifact": "source_aac_copy / 8.9 MB",
      "speakers": "7",
      "turns": "106",
      "switches": "90",
      "unknown": "2"
    },
    {
      "config": "pyannote auto audio shared MIG",
      "duration": "9.6m",
      "asr_time,rtf": "20.8s,0.0362",
      "diarization_time,rtf": "32.3s,0.0562",
      "total_rtf": "0.0993",
      "artifact": "source_aac_copy / 8.9 MB",
      "speakers": "7",
      "turns": "106",
      "switches": "90",
      "unknown": "2"
    },
    {
      "config": "pyannote AAC transcode shared MIG",
      "duration": "9.6m",
      "asr_time,rtf": "19.7s,0.0343",
      "diarization_time,rtf": "30.6s,0.0531",
      "total_rtf": "0.1071",
      "artifact": "aac_transcode / 4.5 MB",
      "speakers": "7",
      "turns": "106",
      "switches": "90",
      "unknown": "2"
    },
    {
      "config": "pyannote auto audio hints 2-8 shared MIG",
      "duration": "1.79h",
      "asr_time,rtf": "1.7m,0.0158",
      "diarization_time,rtf": "5.4m,0.0505",
      "total_rtf": "0.0861",
      "artifact": "aac_transcode / 47.0 MB",
      "speakers": "8",
      "turns": "614",
      "switches": "298",
      "unknown": "24"
    },
    {
      "config": "pyannote auto audio shared MIG",
      "duration": "1.79h",
      "asr_time,rtf": "2.0m,0.0189",
      "diarization_time,rtf": "5.5m,0.0514",
      "total_rtf": "0.0891",
      "artifact": "aac_transcode / 47.0 MB",
      "speakers": "8",
      "turns": "614",
      "switches": "298",
      "unknown": "24"
    },
    {
      "config": "pyannote AAC transcode shared MIG",
      "duration": "1.79h",
      "asr_time,rtf": "1.7m,0.0156",
      "diarization_time,rtf": "5.7m,0.0529",
      "total_rtf": "0.0881",
      "artifact": "aac_transcode / 47.0 MB",
      "speakers": "8",
      "turns": "614",
      "switches": "298",
      "unknown": "24"
    },
    {
      "config": "pyannote auto audio hints 2-8 shared MIG",
      "duration": "9.6m",
      "asr_time,rtf": "28.4s,0.0494",
      "diarization_time,rtf": "35.4s,0.0616",
      "total_rtf": "0.1198",
      "artifact": "source_aac_copy / 8.9 MB",
      "speakers": "7",
      "turns": "106",
      "switches": "90",
      "unknown": "2"
    },
    {
      "config": "pyannote auto audio shared MIG",
      "duration": "9.6m",
      "asr_time,rtf": "17.6s,0.0306",
      "diarization_time,rtf": "35.6s,0.0619",
      "total_rtf": "0.0990",
      "artifact": "source_aac_copy / 8.9 MB",
      "speakers": "7",
      "turns": "106",
      "switches": "90",
      "unknown": "2"
    },
    {
      "config": "pyannote AAC transcode shared MIG",
      "duration": "9.6m",
      "asr_time,rtf": "29.6s,0.0515",
      "diarization_time,rtf": "30.3s,0.0526",
      "total_rtf": "0.1235",
      "artifact": "aac_transcode / 4.5 MB",
      "speakers": "7",
      "turns": "106",
      "switches": "90",
      "unknown": "2"
    },
    {
      "config": "pyannote auto audio shared MIG",
      "duration": "1.79h",
      "asr_time,rtf": "3.9m,0.0362",
      "diarization_time,rtf": "6.3m,0.0588",
      "total_rtf": "0.1113",
      "artifact": "source_aac_copy / 101.2 MB",
      "speakers": "9",
      "turns": "609",
      "switches": "301",
      "unknown": "21"
    },
    {
      "config": "pyannote AAC transcode shared MIG",
      "duration": "1.79h",
      "asr_time,rtf": "1.9m,0.0175",
      "diarization_time,rtf": "5.7m,0.0529",
      "total_rtf": "0.0931",
      "artifact": "aac_transcode / 47.0 MB",
      "speakers": "8",
      "turns": "614",
      "switches": "298",
      "unknown": "24"
    },
    {
      "config": "pyannote auto audio hints 2-8 shared MIG",
      "duration": "9.6m",
      "asr_time,rtf": "15.5s,0.0269",
      "diarization_time,rtf": "30.4s,0.0529",
      "total_rtf": "0.0876",
      "artifact": "source_aac_copy / 8.9 MB",
      "speakers": "7",
      "turns": "106",
      "switches": "90",
      "unknown": "2"
    },
    {
      "config": "pyannote auto audio shared MIG",
      "duration": "9.6m",
      "asr_time,rtf": "16.4s,0.0286",
      "diarization_time,rtf": "31.6s,0.0549",
      "total_rtf": "0.0912",
      "artifact": "source_aac_copy / 8.9 MB",
      "speakers": "7",
      "turns": "106",
      "switches": "90",
      "unknown": "2"
    },
    {
      "config": "pyannote AAC transcode shared MIG",
      "duration": "9.6m",
      "asr_time,rtf": "17.6s,0.0307",
      "diarization_time,rtf": "44.2s,0.0769",
      "total_rtf": "0.1321",
      "artifact": "aac_transcode / 4.5 MB",
      "speakers": "7",
      "turns": "106",
      "switches": "90",
      "unknown": "2"
    }
  ]
}