speaker_segmenter/pyannote-v3-pro/W8A16/SpeakerSegmenter.mlmodelc/metadata.json

[
  {
    "metadataOutputVersion" : "3.0",
    "storagePrecision" : "Mixed (Float16, Palettized (8 bits))",
    "outputSchema" : [
      {
        "hasShapeFlexibility" : "0",
        "isOptional" : "0",
        "dataType" : "Float16",
        "formattedType" : "MultiArray (Float16 21 × 589 × 3)",
        "shortDescription" : "",
        "shape" : "[21, 589, 3]",
        "name" : "speaker_probs",
        "type" : "MultiArray"
      },
      {
        "hasShapeFlexibility" : "0",
        "isOptional" : "0",
        "dataType" : "Float16",
        "formattedType" : "MultiArray (Float16 21 × 589 × 3)",
        "shortDescription" : "",
        "shape" : "[21, 589, 3]",
        "name" : "speaker_ids",
        "type" : "MultiArray"
      },
      {
        "hasShapeFlexibility" : "0",
        "isOptional" : "0",
        "dataType" : "Float16",
        "formattedType" : "MultiArray (Float16 21 × 3)",
        "shortDescription" : "",
        "shape" : "[21, 3]",
        "name" : "speaker_activity",
        "type" : "MultiArray"
      },
      {
        "hasShapeFlexibility" : "0",
        "isOptional" : "0",
        "dataType" : "Float16",
        "formattedType" : "MultiArray (Float16 21 × 589)",
        "shortDescription" : "",
        "shape" : "[21, 589]",
        "name" : "overlapped_speaker_activity",
        "type" : "MultiArray"
      },
      {
        "hasShapeFlexibility" : "0",
        "isOptional" : "0",
        "dataType" : "Float16",
        "formattedType" : "MultiArray (Float16 1767)",
        "shortDescription" : "",
        "shape" : "[1767]",
        "name" : "voice_activity",
        "type" : "MultiArray"
      },
      {
        "hasShapeFlexibility" : "0",
        "isOptional" : "0",
        "dataType" : "Float16",
        "formattedType" : "MultiArray (Float16 21 × 1 × 160000)",
        "shortDescription" : "",
        "shape" : "[21, 1, 160000]",
        "name" : "sliding_window_waveform",
        "type" : "MultiArray"
      }
    ],
    "modelParameters" : [

    ],
    "specificationVersion" : 8,
    "mlProgramOperationTypeHistogram" : {
      "Ios17.reduceArgmax" : 1,
      "Ios16.maxPool" : 3,
      "Ios17.slidingWindows" : 1,
      "Ios17.instanceNorm" : 4,
      "Ios17.exp" : 1,
      "Ios16.softmax" : 1,
      "Ios17.scatter" : 42,
      "Ios17.transpose" : 2,
      "Ios17.expandDims" : 1,
      "Ios16.reduceMax" : 1,
      "Ios17.add" : 40,
      "Ios17.sliceByIndex" : 61,
      "Ios16.reduceSum" : 2,
      "Ios17.log" : 1,
      "Ios17.conv" : 3,
      "Ios17.sub" : 1,
      "Ios16.constexprLutToDense" : 22,
      "Ios17.lstm" : 4,
      "OneHot" : 1,
      "Ios17.linear" : 5,
      "Ios17.leakyRelu" : 5,
      "Ios17.abs" : 1,
      "Ios17.cast" : 2,
      "Ios17.realDiv" : 1,
      "Ios17.greater" : 1,
      "Ios17.mul" : 1
    },
    "computePrecision" : "Mixed (Float16, Float32, 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 480000)",
        "shortDescription" : "",
        "shape" : "[480000]",
        "name" : "waveform",
        "type" : "MultiArray"
      },
      {
        "hasShapeFlexibility" : "0",
        "isOptional" : "0",
        "dataType" : "Float16",
        "formattedType" : "MultiArray (Float16 7)",
        "shortDescription" : "",
        "shape" : "[7]",
        "name" : "input_1",
        "type" : "MultiArray"
      }
    ],
    "generatedClassName" : "SpeakerSegmenter_8_bit",
    "method" : "predict"
  }
]