[
  {
    "metadataOutputVersion" : "3.0",
    "shortDescription" : "emotion2vec_plus_base speech emotion recognition (9 classes). Input: 16 kHz mono float32 waveform, 0.5-10 s.",
    "outputSchema" : [
      {
        "hasShapeFlexibility" : "0",
        "isOptional" : "0",
        "dataType" : "Float32",
        "formattedType" : "MultiArray (Float32 1 × 9)",
        "shortDescription" : "Softmax probabilities [1, 9]: 0=angry, 1=disgusted, 2=fearful, 3=happy, 4=neutral, 5=other, 6=sad, 7=surprised, 8=unknown",
        "shape" : "[1, 9]",
        "name" : "probs",
        "type" : "MultiArray"
      },
      {
        "hasShapeFlexibility" : "0",
        "isOptional" : "0",
        "dataType" : "Float32",
        "formattedType" : "MultiArray (Float32 1 × 9)",
        "shortDescription" : "Pre-softmax logits [1, 9]",
        "shape" : "[1, 9]",
        "name" : "logits",
        "type" : "MultiArray"
      }
    ],
    "storagePrecision" : "Mixed (Float16, Float32)",
    "modelParameters" : [

    ],
    "author" : "emotion2vec (Ma et al.), converted from FunASR checkpoint emotion2vec\/emotion2vec_plus_base",
    "specificationVersion" : 9,
    "license" : "See emotion2vec\/emotion2vec_plus_base model card",
    "mlProgramOperationTypeHistogram" : {
      "Ios18.expandDims" : 7,
      "Ios18.mul" : 18,
      "Ios18.softmax" : 13,
      "Ios18.rsqrt" : 1,
      "Ios18.matmul" : 24,
      "Ios16.reduceMean" : 3,
      "Ios18.greaterEqual" : 1,
      "Ios18.add" : 38,
      "Ios18.layerNorm" : 38,
      "Ios16.fillLike" : 1,
      "Ios18.reshape" : 24,
      "Ios18.linear" : 50,
      "Ios18.conv" : 12,
      "Ios18.gelu" : 24,
      "Ios18.sub" : 3,
      "Ios18.concat" : 1,
      "Ios18.cast" : 8,
      "Ios18.transpose" : 47,
      "Ios18.abs" : 1,
      "Ios18.sliceByIndex" : 39,
      "Ios16.cumsum" : 1
    },
    "computePrecision" : "Mixed (Float16, Float32, Int32)",
    "stateSchema" : [

    ],
    "isUpdatable" : "0",
    "availability" : {
      "macOS" : "15.0",
      "tvOS" : "18.0",
      "visionOS" : "2.0",
      "watchOS" : "11.0",
      "iOS" : "18.0",
      "macCatalyst" : "18.0"
    },
    "modelType" : {
      "name" : "MLModelType_mlProgram"
    },
    "inputSchema" : [
      {
        "dataType" : "Float32",
        "hasShapeFlexibility" : "1",
        "isOptional" : "0",
        "shapeFlexibility" : "1 × 8000...160000",
        "shapeRange" : "[[1, 1], [8000, 160000]]",
        "formattedType" : "MultiArray (Float32 1 × 48000)",
        "type" : "MultiArray",
        "shape" : "[1, 48000]",
        "name" : "audio",
        "shortDescription" : "Raw mono waveform at 16 kHz, float32 in [-1, 1], shape [1, N], 8000 <= N <= 160000. Normalised internally."
      }
    ],
    "userDefinedMetadata" : {
      "labels_original" : "生气\/angry,厌恶\/disgusted,恐惧\/fearful,开心\/happy,中立\/neutral,其他\/other,难过\/sad,吃惊\/surprised,<unk>",
      "com.github.apple.coremltools.source_dialect" : "TorchScript",
      "com.github.apple.coremltools.conversion_date" : "2026-10-06",
      "labels" : "angry,disgusted,fearful,happy,neutral,other,sad,surprised,unknown",
      "com.github.apple.coremltools.source" : "torch==2.7.0",
      "sample_rate" : "16000",
      "com.github.apple.coremltools.version" : "9.0"
    },
    "generatedClassName" : "Emotion2Vec",
    "method" : "predict"
  }
]