{
  "schema_version": 1,
  "family": "crisperwhisper",
  "display_name": "CrisperWhisper 2.0",
  "description": "Whisper-based verbatim and intended speech transcription.",
  "category": "asr",
  "status": "supported",
  "tasks": [
    "asr",
    "align"
  ],
  "modes": [
    "offline",
    "streaming"
  ],
  "languages": [
    "af",
    "am",
    "ar",
    "as",
    "az",
    "ba",
    "be",
    "bg",
    "bn",
    "bo",
    "br",
    "bs",
    "ca",
    "cs",
    "cy",
    "da",
    "de",
    "el",
    "en",
    "es",
    "et",
    "eu",
    "fa",
    "fi",
    "fo",
    "fr",
    "gl",
    "gu",
    "ha",
    "haw",
    "he",
    "hi",
    "hr",
    "ht",
    "hu",
    "hy",
    "id",
    "is",
    "it",
    "ja",
    "jw",
    "ka",
    "kk",
    "km",
    "kn",
    "ko",
    "la",
    "lb",
    "ln",
    "lo",
    "lt",
    "lv",
    "mg",
    "mi",
    "mk",
    "ml",
    "mn",
    "mr",
    "ms",
    "mt",
    "my",
    "ne",
    "nl",
    "nn",
    "no",
    "oc",
    "pa",
    "pl",
    "ps",
    "pt",
    "ro",
    "ru",
    "sa",
    "sd",
    "si",
    "sk",
    "sl",
    "sn",
    "so",
    "sq",
    "sr",
    "su",
    "sv",
    "sw",
    "ta",
    "te",
    "tg",
    "th",
    "tk",
    "tl",
    "tr",
    "tt",
    "uk",
    "ur",
    "uz",
    "vi",
    "yi",
    "yo",
    "zh"
  ],
  "capabilities": {
    "align": [
      "word_timestamps"
    ],
    "asr": [
      "word_timestamps",
      "partial_results"
    ]
  },
  "dependencies": [],
  "options": {
    "request": [
      {
        "name": "return_timestamps",
        "required": false,
        "type": "bool",
        "description": "Return word-level timestamps.",
        "default": false
      },
      {
        "name": "audio_chunk_mode",
        "required": false,
        "type": "enum",
        "values": [
          "auto",
          "fixed",
          "none"
        ],
        "description": "Continuation window planning. None requires audio no longer than 30 seconds.",
        "default": "auto"
      },
      {
        "name": "audio_chunk_duration_sec",
        "required": false,
        "type": "float",
        "description": "Duration of each encoder window in seconds. Must be positive and greater than overlap.",
        "max": 30,
        "default": 30
      },
      {
        "name": "audio_chunk_overlap_sec",
        "required": false,
        "type": "float",
        "description": "Audio overlap between continuation windows in seconds. Must be smaller than chunk duration.",
        "min": 0,
        "default": 4
      },
      {
        "name": "language",
        "required": false,
        "type": "string",
        "description": "Audio language code.",
        "default": "en"
      },
      {
        "name": "transcription_mode",
        "required": false,
        "type": "enum",
        "description": "Verbatim speech, intended clean text, or transcript-conditioned verbatimize.",
        "values": [
          "verbatim",
          "intended",
          "verbatimize"
        ],
        "default": "verbatim"
      },
      {
        "name": "max_tokens",
        "required": false,
        "type": "int",
        "description": "Maximum generated tokens per window. Positive budget, additionally bounded by remaining checkpoint text positions after the prompt.",
        "min": 1,
        "default": 256
      }
    ],
    "session": [
      {
        "name": "weight_type",
        "required": false,
        "type": "enum",
        "values": [
          "native",
          "f32",
          "f16",
          "bf16",
          "q8_0",
          "q4_0",
          "q4_1",
          "q5_0",
          "q5_1",
          "q2_k",
          "q3_k",
          "q4_k",
          "q5_k",
          "q6_k",
          "i8"
        ],
        "description": "Matmul weight storage type.",
        "default": "native"
      }
    ],
    "load": []
  },
  "runtime": {
    "tags": [
      "gguf"
    ]
  },
  "ui": {
    "tags": [
      "ASR",
      "GGUF"
    ],
    "docs": [
      "docs/models/crisperwhisper.md"
    ],
    "recommended_package": "crisperwhisper_2_large_q8_0"
  },
  "package_defaults": {
    "download": {
      "kind": "huggingface_snapshot",
      "repo": "audio-cpp/audio.cpp-gguf",
      "revision": "main",
      "gated": false
    }
  },
  "packages": [
    {
      "id": "crisperwhisper_2_large_bf16",
      "display_name": "CrisperWhisper 2.0 Large GGUF BF16",
      "format": "gguf",
      "precision": "bf16",
      "target_directory": "CrisperWhisper2.0-GGUF",
      "files": [
        "CrisperWhisper2.0-GGUF/crisperwhisper-2-large-bf16.gguf"
      ],
      "strip_prefix": "CrisperWhisper2.0-GGUF"
    },
    {
      "id": "crisperwhisper_2_large_q8_0",
      "display_name": "CrisperWhisper 2.0 Large GGUF Q8_0",
      "default": true,
      "format": "gguf",
      "precision": "q8_0",
      "target_directory": "CrisperWhisper2.0-GGUF",
      "files": [
        "CrisperWhisper2.0-GGUF/crisperwhisper-2-large-q8_0.gguf"
      ],
      "strip_prefix": "CrisperWhisper2.0-GGUF"
    },
    {
      "id": "crisperwhisper_2_large_q4_k",
      "display_name": "CrisperWhisper 2.0 Large GGUF Q4_K",
      "format": "gguf",
      "precision": "q4_k",
      "target_directory": "CrisperWhisper2.0-GGUF",
      "files": [
        "CrisperWhisper2.0-GGUF/crisperwhisper-2-large-q4_k.gguf"
      ],
      "strip_prefix": "CrisperWhisper2.0-GGUF"
    }
  ],
  "sources": [
    {
      "format": "safetensors",
      "roots": {
        "model": "."
      },
      "files": {
        "config": "model:config.json",
        "generation_config": "model:generation_config.json",
        "tokenizer": "model:tokenizer.json",
        "tokenizer_config": "model:tokenizer_config.json",
        "vocab": "model:vocab.json",
        "merges": "model:merges.txt"
      },
      "tensors": {
        "weights": "model:model.safetensors"
      }
    },
    {
      "format": "gguf",
      "roots": {
        "model": ".",
        "weights": "$gguf"
      },
      "files": {
        "config": "model:config.json",
        "generation_config": "model:generation_config.json",
        "tokenizer": "model:tokenizer.json",
        "tokenizer_config": "model:tokenizer_config.json",
        "vocab": "model:vocab.json",
        "merges": "model:merges.txt"
      },
      "tensors": {
        "weights": "weights:"
      }
    }
  ]
}