voice_audio_multi/model_specs/kitten_tts.json

{
  "schema_version": 1,
  "family": "kitten_tts",
  "display_name": "KittenTTS Mini 0.8",
  "description": "Native KittenTTS Mini 0.8 English text-to-speech with eight built-in voices and shared eSpeak-ng phonemization.",
  "category": "tts",
  "status": "community",
  "tasks": [
    "tts"
  ],
  "modes": [
    "offline"
  ],
  "languages": [
    "en"
  ],
  "runtime": {
    "tags": [
      "gguf"
    ]
  },
  "capabilities": {
    "tts": [
      "built_in_voices",
      "long_form"
    ]
  },
  "options": {
    "request": [
      {
        "name": "voice_id",
        "type": "enum",
        "values": ["Bella","Bruno","Hugo","Jasper","Kiki","Leo","Luna","Rosie","expr-voice-2-f","expr-voice-2-m","expr-voice-3-f","expr-voice-3-m","expr-voice-4-f","expr-voice-4-m","expr-voice-5-f","expr-voice-5-m"],
        "description": "Built-in voice ID or package alias; structured voice input (CLI --voice-id) takes precedence.",
        "required": false,
        "default": "expr-voice-5-m"
      },
      {
        "name": "speed",
        "type": "float",
        "description": "Speech speed multiplier; larger values produce faster speech. Must be finite and positive.",
        "required": false,
        "default": 1
      },
      {
        "name": "seed",
        "type": "int",
        "description": "Decoder noise seed for the current request; omitted requests choose a fresh random seed.",
        "required": false,
        "min": 0
      },
      {
        "name": "text_chunk_mode",
        "type": "enum",
        "description": "Long-form text chunking mode.",
        "values": [
          "default",
          "tag_aware",
          "japanese",
          "endline"
        ],
        "required": false,
        "default": "default"
      },
      {
        "name": "text_chunk_size",
        "type": "int",
        "description": "Maximum Unicode codepoints per long-form chunk.",
        "required": false,
        "min": 1,
        "default": 400
      }
    ],
    "session": [
      {
        "name": "weight_type",
        "type": "enum",
        "preset": "weight_type_full",
        "description": "Storage type for matrix multiplication weights.",
        "required": false,
        "default": "native"
      },
      {
        "name": "conv_weight_type",
        "type": "enum",
        "preset": "weight_type_conv",
        "description": "Storage type for convolution weights.",
        "required": false,
        "default": "native"
      },
      {
        "name": "graph_capacity_mode",
        "type": "enum",
        "description": "Offline graph capacity policy.",
        "values": [
          "fixed",
          "tiered",
          "grow",
          "double"
        ],
        "required": false,
        "default": "fixed"
      },
      {
        "name": "max_input_tokens",
        "type": "int",
        "description": "Maximum predictor input token capacity; defaults to min(512, checkpoint context_length).",
        "required": false,
        "min": 1,
        "max": 512,
        "default": 512
      },
      {
        "name": "pre_tail_tokens",
        "type": "int",
        "description": "Optional prebuilt predictor tail token capacity.",
        "required": false,
        "min": 0,
        "max": 512,
        "default": 0
      },
      {
        "name": "espeak_library_path",
        "type": "path",
        "description": "Optional explicit path to the eSpeak-ng shared library.",
        "required": false
      },
      {
        "name": "espeak_data_path",
        "type": "path",
        "description": "Optional explicit path to espeak-ng-data.",
        "required": false
      }
    ],
    "load": []
  },
  "package_defaults": {
    "download": {
      "kind": "huggingface_snapshot",
      "repo": "audio-cpp/audio.cpp-gguf",
      "revision": "main",
      "gated": false
    }
  },
  "packages": [
    {
      "id": "kitten_tts_mini_0_8_orig",
      "display_name": "KittenTTS Mini 0.8 Original-Dtype GGUF",
      "default": true,
      "format": "gguf",
      "precision": "orig",
      "target_directory": "KittenTTS-GGUF",
      "files": [
        "KittenTTS-GGUF/kitten-tts-mini-0.8-orig.gguf"
      ],
      "strip_prefix": "KittenTTS-GGUF"
    }
  ],
  "dependencies": [],
  "ui": {
    "recommended_package": "kitten_tts_mini_0_8_orig",
    "tags": [
      "TTS",
      "GGUF"
    ],
    "builtin_voices": [
      "Bella",
      "Jasper",
      "Luna",
      "Bruno",
      "Rosie",
      "Hugo",
      "Kiki",
      "Leo"
    ],
    "default_voice": "Leo",
    "docs": [
      "docs/models/kitten_tts.md",
      "docs/tts.md",
      "docs/gguf.md"
    ]
  },
  "sources": [
    {
      "format": "gguf",
      "roots": {
        "model": ".",
        "weights": "$gguf"
      },
      "files": {
        "config": "model:config.json"
      },
      "tensors": {
        "weights": {
          "source": "weights:",
          "prefix": "kitten"
        }
      }
    },
    {
      "format": "safetensors",
      "roots": {
        "model": "."
      },
      "files": {
        "config": "model:config.json"
      },
      "tensors": {
        "weights": "model:ggml/kitten_tts.safetensors"
      }
    }
  ]
}