{
  "recorded_on": "2026-10-13",
  "captured_at": "2026-10-13T14:09:11.160314+00:00",
  "method": "Unmodified downloadable whisper-transcription-v1 CLI, sequential English then Russian",
  "scope": "Real GPU transcription with the downloadable CLI. Not a completed browser WebUI test. Inputs are short clean synthesized clips, not an interview/noise benchmark.",
  "gpu": "NVIDIA GeForce RTX 4090, 24564 MiB, 610.43.02",
  "python": "3.11.16 (main, Sep  1 2026, 14:18:37) [Clang 22.1.3 ]",
  "packages": {
    "faster-whisper": "1.2.1",
    "ctranslate2": "4.6.0",
    "av": "16.0.1",
    "huggingface-hub": "0.36.0",
    "tokenizers": "0.22.1",
    "onnxruntime": "1.23.2",
    "numpy": "2.2.6"
  },
  "model": {
    "repository": "Systran/faster-whisper-large-v3",
    "revision": "edaa852ec7e145841d8ffdb056a99866b5f0a478"
  },
  "settings": {
    "example_version": 1,
    "model": {
      "repository": "Systran/faster-whisper-large-v3",
      "revision": "edaa852ec7e145841d8ffdb056a99866b5f0a478"
    },
    "device": "cuda",
    "compute_type": "float16",
    "language": "auto",
    "beam_size": 5,
    "cpu_threads": 4,
    "vad_filter": true,
    "task": "transcribe",
    "versions": {
      "faster-whisper": "1.2.1",
      "ctranslate2": "4.6.0",
      "av": "16.0.1",
      "onnxruntime": "1.23.2",
      "tokenizers": "0.22.1",
      "numpy": "2.2.6",
      "huggingface-hub": "0.36.0"
    }
  },
  "model_load_seconds": 2.878180107101798,
  "timing_note": "One shared model load before the first input; processing_seconds excludes model download, model load, environment setup and upload. The second file reuses the loaded model. Single measurements, not comparative GPU benchmarks.",
  "source_example_sha256": "d50f467fca73bf2fde2f0106d5690c4bb3b220bc2ad03e1a2cde490f81aa4403",
  "source_audio_manifest": "application-assets/tts/samples-20261013.json",
  "command": "python transcribe.py --input input --output output --cache models --model large-v3 --device cuda --compute-type float16 --language auto --offline",
  "observed_limitations": [
    "English transcript omitted source punctuation and capitalization.",
    "English final raw segment ends at 15.92 seconds, beyond the 8.32-second input. Raw TXT/SRT/VTT/JSON exports are preserved without corrections."
  ],
  "samples": [
    {
      "language": "en",
      "input": {
        "file": "application-assets/tts/chatterbox-multilingual-en-20261013.wav",
        "source_script": "Welcome to the demo. Today we will turn a short script into a voiceover, listen to the result, and download the clip for editing.",
        "sha256": "26cfae91a4a6357bf9c93a556a3dd0927eeda4d44aaf28e6da7314c66dc7593b",
        "bytes": 399404,
        "duration_seconds": 8.32,
        "source": "Chatterbox Multilingual synthetic speech, Emily.wav predefined voice; captured 2026-10-13"
      },
      "transcript": "welcome to the demo today we will turn a short script into a voiceover\nlisten to the result and download the clip for editing\n",
      "detected_language": "en",
      "processing_seconds": 0.9912004470825195,
      "segments": [
        {
          "start": 0.0,
          "end": 4.0,
          "text": "welcome to the demo today we will turn a short script into a voiceover"
        },
        {
          "start": 4.0,
          "end": 15.92,
          "text": "listen to the result and download the clip for editing"
        }
      ],
      "files": {
        "transcript.txt": "en/transcript.txt",
        "transcript.srt": "en/transcript.srt",
        "transcript.vtt": "en/transcript.vtt",
        "transcript.json": "en/transcript.json",
        "complete.json": "en/complete.json"
      },
      "sha256": {
        "transcript.txt": "db29e1b5d292198573f324112954800ae4cc8beb19ac1c964ef39cb332ed6629",
        "transcript.json": "10459fde10e8011e033d39bf293dae7dbae3e8de2a0bbb9ce76c2a325e519585",
        "transcript.vtt": "904a08755fcb4bf0e7d964cf45a478dd8811cc96d5e819a57081c135cf4bacb7",
        "transcript.srt": "b3ca2c9f79224625ea1964b891bd3e43706e7db62f05c66542a5e6b4367444e0",
        "complete.json": "62a4a68e2f46b9ec0c8b1f66b01b9696f339ff453c5614819600dbafc5a6958a"
      }
    },
    {
      "language": "ru",
      "input": {
        "file": "application-assets/tts/chatterbox-multilingual-ru-20261013.wav",
        "source_script": "Добро пожаловать на демонстрацию. Сегодня мы превратим короткий сценарий в озвучку, послушаем результат и скачаем файл для монтажа.",
        "sha256": "5772e5688f7e8d37cf418100f272e7b541267bb60a452aea34c3a60726e33983",
        "bytes": 399404,
        "duration_seconds": 8.32,
        "source": "Chatterbox Multilingual synthetic speech, Emily.wav predefined voice; captured 2026-10-13"
      },
      "transcript": "Добро пожаловать на демонстрацию.\nСегодня мы превратим короткий сценарий в озвучку,\nпослушаем результат и скачаем файл для монтажа.\n",
      "detected_language": "ru",
      "processing_seconds": 0.628773215925321,
      "segments": [
        {
          "start": 0.0,
          "end": 1.92,
          "text": "Добро пожаловать на демонстрацию."
        },
        {
          "start": 2.16,
          "end": 4.58,
          "text": "Сегодня мы превратим короткий сценарий в озвучку,"
        },
        {
          "start": 4.84,
          "end": 7.5,
          "text": "послушаем результат и скачаем файл для монтажа."
        }
      ],
      "files": {
        "transcript.txt": "ru/transcript.txt",
        "transcript.srt": "ru/transcript.srt",
        "transcript.vtt": "ru/transcript.vtt",
        "transcript.json": "ru/transcript.json",
        "complete.json": "ru/complete.json"
      },
      "sha256": {
        "transcript.txt": "e7050e91940471002a19f3c1a141f6e861da59b764b6fc9d88b7f12344b36ea1",
        "transcript.json": "d4234a50a0a997361b8a058444c813f691b0ca9d289564c2d814f5615de26725",
        "transcript.vtt": "a29a30c0f06d86276ca00827d38c2544648911c201e32955420f7b55627b59f8",
        "transcript.srt": "0fc62549366df064000b35bc10bdf70cfaefd6c5733dc83dd5fdce904da56e18",
        "complete.json": "74cc7795f1ac5ad613ea43ee5312913485131684cd214a0131f3305ddd20f815"
      }
    }
  ]
}