[ { "shortDescription" : "VAD with SE modules trained on MUSAN (86.47% accuracy)", "metadataOutputVersion" : "3.0", "outputSchema" : [ { "hasShapeFlexibility" : "0", "isOptional" : "0", "dataType" : "Float32", "formattedType" : "MultiArray (Float32 1 × 1)", "shortDescription" : "", "shape" : "[1, 1]", "name" : "vad_probability", "type" : "MultiArray" } ], "version" : "2.0", "modelParameters" : [ ], "author" : "Silero VAD with Trained SE Modules", "specificationVersion" : 7, "storagePrecision" : "Mixed (Float16, Float32, Palettized (4 bits))", "mlProgramOperationTypeHistogram" : { "Concat" : 4, "Ios16.cast" : 4, "Ios16.mul" : 5, "Ios16.layerNorm" : 1, "SliceByIndex" : 3, "Ios16.constexprLutToDense" : 8, "Ios16.sqrt" : 3, "Stack" : 1, "Transpose" : 3, "Ios16.sigmoid" : 5, "Ios16.conv" : 5, "Ios16.add" : 6, "Ios16.linear" : 14, "Ios16.reduceMean" : 5, "Ios16.relu" : 9, "Ios16.lstm" : 1, "Ios16.pow" : 6, "Squeeze" : 1, "Ios16.reshape" : 8 }, "computePrecision" : "Mixed (Float16, Float32, Int32)", "stateSchema" : [ ], "isUpdatable" : "0", "availability" : { "macOS" : "13.0", "tvOS" : "16.0", "visionOS" : "1.0", "watchOS" : "9.0", "iOS" : "16.0", "macCatalyst" : "16.0" }, "modelType" : { "name" : "MLModelType_mlProgram" }, "inputSchema" : [ { "hasShapeFlexibility" : "0", "isOptional" : "0", "dataType" : "Float32", "formattedType" : "MultiArray (Float32 1 × 512)", "shortDescription" : "", "shape" : "[1, 512]", "name" : "audio_chunk", "type" : "MultiArray" } ], "userDefinedMetadata" : { "com.github.apple.coremltools.source_dialect" : "TorchScript", "com.github.apple.coremltools.version" : "8.3.0", "com.github.apple.coremltools.source" : "torch==2.5.0" }, "generatedClassName" : "silero_vad_se_trained_palettize_4bit", "method" : "predict" } ]