voice_audio_multi/model_specs/miocodec.json

{
  "family": "miocodec",
  "display_name": "MioCodec",
  "description": "Lightweight neural audio codec that encodes 44.1 kHz audio into a low-rate 25 Hz discrete token stream and decodes tokens with speaker conditioning for efficient speech modeling.",
  "category": "audio_tools",
  "status": "supported",
  "tasks": [
    "vc",
    "s2s"
  ],
  "modes": [
    "offline"
  ],
  "languages": [
    "language_agnostic"
  ],
  "options": {
    "request": [],
    "session": [
      {
        "name": "weight_type",
        "type": "enum",
        "description": "Weight storage type.",
        "required": false,
        "preset": "weight_type_full",
        "default": "f32"
      },
      {
        "name": "weight_context_mb",
        "type": "int",
        "description": "Context or graph arena size in MiB.",
        "required": false,
        "min": 1,
        "default": 256
      },
      {
        "name": "constant_context_mb",
        "type": "int",
        "description": "Context or graph arena size in MiB.",
        "required": false,
        "min": 1,
        "default": 256
      },
      {
        "name": "content_graph_arena_mb",
        "type": "int",
        "description": "Context or graph arena size in MiB.",
        "required": false,
        "min": 1,
        "default": 512
      },
      {
        "name": "global_graph_arena_mb",
        "type": "int",
        "description": "Context or graph arena size in MiB.",
        "required": false,
        "min": 1,
        "default": 256
      },
      {
        "name": "wave_graph_arena_mb",
        "type": "int",
        "description": "Context or graph arena size in MiB.",
        "required": false,
        "min": 1,
        "default": 512
      }
    ],
    "load": []
  },
  "capabilities": {
    "codec": [
      "encode_decode"
    ],
    "vc": [
      "speaker_reference"
    ],
    "s2s": [
      "speaker_reference"
    ]
  },
  "runtime": {
    "tags": [
      "gguf"
    ]
  },
  "ui": {
    "recommended_package": "miocodec_q8_0",
    "tags": [
      "Codec",
      "VC",
      "GGUF"
    ],
    "docs": [
      "docs/audio_tools.md",
      "docs/gguf.md"
    ]
  },
  "package_defaults": {
    "download": {
      "kind": "huggingface_snapshot",
      "repo": "audio-cpp/audio.cpp-gguf",
      "revision": "main",
      "gated": false
    }
  },
  "packages": [
    {
      "id": "miocodec_q8_0",
      "display_name": "MioCodec 25Hz 44.1kHz v2 Q8_0 GGUF",
      "default": true,
      "format": "gguf",
      "precision": "q8_0",
      "target_directory": "MioCodec-25Hz-44.1kHz-v2-GGUF",
      "files": [
        "MioCodec-25Hz-44.1kHz-v2-GGUF/miocodec-25hz-44khz-v2-q8_0.gguf"
      ],
      "strip_prefix": "MioCodec-25Hz-44.1kHz-v2-GGUF"
    },
    {
      "id": "miocodec_f16",
      "display_name": "MioCodec 25Hz 44.1kHz v2 F16 GGUF",
      "format": "gguf",
      "precision": "f16",
      "target_directory": "MioCodec-25Hz-44.1kHz-v2-GGUF",
      "files": [
        "MioCodec-25Hz-44.1kHz-v2-GGUF/miocodec-25hz-44khz-v2-f16.gguf"
      ],
      "strip_prefix": "MioCodec-25Hz-44.1kHz-v2-GGUF"
    },
    {
      "id": "miocodec_orig",
      "display_name": "MioCodec 25Hz 44.1kHz v2 Original-Dtype GGUF",
      "format": "gguf",
      "precision": "orig",
      "target_directory": "MioCodec-25Hz-44.1kHz-v2-GGUF",
      "files": [
        "MioCodec-25Hz-44.1kHz-v2-GGUF/miocodec-25hz-44khz-v2-orig.gguf"
      ],
      "strip_prefix": "MioCodec-25Hz-44.1kHz-v2-GGUF"
    }
  ],
  "sources": [
    {
      "format": "gguf",
      "roots": {
        "model": ".",
        "weights": "$gguf"
      },
      "files": {
        "config": "model:config.yaml"
      },
      "tensors": {
        "weights": {
          "source": "weights:",
          "prefix": "weights"
        },
        "wavlm_weights": {
          "source": "weights:",
          "prefix": "wavlm_weights"
        }
      }
    },
    {
      "format": "safetensors",
      "roots": {
        "model": ".",
        "wavlm": "wavlm-base-plus-mlx"
      },
      "files": {
        "config": "model:config.yaml"
      },
      "tensors": {
        "weights": "model:model.safetensors",
        "wavlm_weights": "wavlm:weights.safetensors"
      }
    }
  ]
}