[
  {
    "metadataOutputVersion" : "3.0",
    "storagePrecision" : "Float16",
    "outputSchema" : [
      {
        "hasShapeFlexibility" : "0",
        "isOptional" : "0",
        "dataType" : "Float16",
        "formattedType" : "MultiArray (Float16 1 × 1024 × 1 × 188)",
        "shortDescription" : "",
        "shape" : "[1, 1024, 1, 188]",
        "name" : "encoder_output_embeds",
        "type" : "MultiArray"
      },
      {
        "hasShapeFlexibility" : "0",
        "isOptional" : "0",
        "dataType" : "Float16",
        "formattedType" : "MultiArray (Float16 1 × 640 × 1 × 188)",
        "shortDescription" : "",
        "shape" : "[1, 640, 1, 188]",
        "name" : "joint_projected_encoder_output_embeds",
        "type" : "MultiArray"
      }
    ],
    "modelParameters" : [

    ],
    "specificationVersion" : 8,
    "mlProgramOperationTypeHistogram" : {
      "Ios16.silu" : 72,
      "Ios17.mul" : 73,
      "Split" : 24,
      "Ios17.transpose" : 1,
      "Ios17.sub" : 1,
      "Ios17.matmul" : 72,
      "Ios17.conv" : 295,
      "Ios16.sigmoid" : 24,
      "Ios17.add" : 168,
      "Ios17.sliceByIndex" : 48,
      "Ios17.batchNorm" : 120,
      "Ios16.relu" : 3,
      "Ios16.softmax" : 24,
      "Ios17.reshape" : 193,
      "Ios17.layerNorm" : 120,
      "Pad" : 24
    },
    "computePrecision" : "Mixed (Float16, Int32)",
    "isUpdatable" : "0",
    "stateSchema" : [

    ],
    "availability" : {
      "macOS" : "14.0",
      "tvOS" : "17.0",
      "visionOS" : "1.0",
      "watchOS" : "10.0",
      "iOS" : "17.0",
      "macCatalyst" : "17.0"
    },
    "modelType" : {
      "name" : "MLModelType_mlProgram"
    },
    "userDefinedMetadata" : {
      "com.github.apple.coremltools.source_dialect" : "TorchScript",
      "com.github.apple.coremltools.source" : "torch==2.6.0",
      "com.github.apple.coremltools.version" : "8.2"
    },
    "inputSchema" : [
      {
        "hasShapeFlexibility" : "0",
        "isOptional" : "0",
        "dataType" : "Float16",
        "formattedType" : "MultiArray (Float16 1 × 1 × 1501 × 128)",
        "shortDescription" : "",
        "shape" : "[1, 1, 1501, 128]",
        "name" : "melspectrogram_features",
        "type" : "MultiArray"
      },
      {
        "hasShapeFlexibility" : "0",
        "isOptional" : "0",
        "dataType" : "Float16",
        "formattedType" : "MultiArray (Float16 1 × 1 × 1 × 1)",
        "shortDescription" : "",
        "shape" : "[1, 1, 1, 1]",
        "name" : "input_1",
        "type" : "MultiArray"
      }
    ],
    "generatedClassName" : "AudioEncoder",
    "method" : "predict"
  }
]