{
  "model_name": "wav2arkit_cpu",
  "description": "End-to-end audio to ARKit blendshape model",
  "format": "onnx",
  
  "source_models": {
    "audio_encoder": "facebook/wav2vec2-base-960h",
    "expression_decoder": "3DAIGC/LAM_audio2exp"
  },

  "audio_encoder": {
    "source": "facebook/wav2vec2-base-960h",
    "hidden_size": 768
  },

  "preprocessing": {
    "sample_rate": 16000,
    "channels": 1,
    "normalize": false
  },

  "input_spec": {
    "name": "audio_waveform",
    "dtype": "float32",
    "shape": ["batch_size", "num_samples"]
  },

  "output_spec": {
    "name": "blendshapes",
    "dtype": "float32",
    "shape": ["batch_size", "num_frames", 52],
    "fps": 30,
    "value_range": [0.0, 1.0]
  },

  "num_blendshapes": 52,
  "output_fps": 30,
  "frame_formula": "ceil(30 * num_samples / 16000)",

  "blendshape_names": [
    "browDownLeft", "browDownRight", "browInnerUp", "browOuterUpLeft", "browOuterUpRight",
    "cheekPuff", "cheekSquintLeft", "cheekSquintRight",
    "eyeBlinkLeft", "eyeBlinkRight", "eyeLookDownLeft", "eyeLookDownRight",
    "eyeLookInLeft", "eyeLookInRight", "eyeLookOutLeft", "eyeLookOutRight",
    "eyeLookUpLeft", "eyeLookUpRight", "eyeSquintLeft", "eyeSquintRight",
    "eyeWideLeft", "eyeWideRight",
    "jawForward", "jawLeft", "jawOpen", "jawRight",
    "mouthClose", "mouthDimpleLeft", "mouthDimpleRight", "mouthFrownLeft", "mouthFrownRight",
    "mouthFunnel", "mouthLeft", "mouthLowerDownLeft", "mouthLowerDownRight",
    "mouthPressLeft", "mouthPressRight", "mouthPucker", "mouthRight",
    "mouthRollLower", "mouthRollUpper", "mouthShrugLower", "mouthShrugUpper",
    "mouthSmileLeft", "mouthSmileRight", "mouthStretchLeft", "mouthStretchRight",
    "mouthUpperUpLeft", "mouthUpperUpRight",
    "noseSneerLeft", "noseSneerRight", "tongueOut"
  ],

  "onnx": {
    "opset_version": 18,
    "producer": "pytorch",
    "model_file": "wav2arkit_cpu.onnx"
  }
}