[ { "metadataOutputVersion" : "3.0", "storagePrecision" : "Float16", "outputSchema" : [ { "hasShapeFlexibility" : "0", "isOptional" : "0", "dataType" : "Float16", "formattedType" : "MultiArray (Float16 1 × 384 × 1 × 1500)", "shortDescription" : "", "shape" : "[1, 384, 1, 1500]", "name" : "encoder_output_embeds", "type" : "MultiArray" }, { "hasShapeFlexibility" : "0", "isOptional" : "0", "dataType" : "Float16", "formattedType" : "MultiArray (Float16 4 × 384 × 1 × 1536)", "shortDescription" : "", "shape" : "[4, 384, 1, 1536]", "name" : "encoder_attn_key_cache", "type" : "MultiArray" }, { "hasShapeFlexibility" : "0", "isOptional" : "0", "dataType" : "Float16", "formattedType" : "MultiArray (Float16 4 × 384 × 1 × 1536)", "shortDescription" : "", "shape" : "[4, 384, 1, 1536]", "name" : "encoder_attn_value_cache", "type" : "MultiArray" } ], "modelParameters" : [ ], "specificationVersion" : 9, "mlProgramOperationTypeHistogram" : { "Pad" : 2, "Ios18.batchNorm" : 9, "Ios18.conv" : 34, "Ios18.gelu" : 6, "Ios18.concat" : 30, "Ios16.einsum" : 192, "Ios18.add" : 9, "Ios18.softmax" : 96, "Ios18.sliceByIndex" : 168, "Ios18.layerNorm" : 9, "Ios18.transpose" : 4, "Ios18.mul" : 96 }, "computePrecision" : "Mixed (Float16, Int32)", "isUpdatable" : "0", "stateSchema" : [ ], "availability" : { "macOS" : "15.0", "tvOS" : "18.0", "visionOS" : "2.0", "watchOS" : "11.0", "iOS" : "18.0", "macCatalyst" : "18.0" }, "modelType" : { "name" : "MLModelType_mlProgram" }, "userDefinedMetadata" : { "com.github.apple.coremltools.source_dialect" : "TorchScript", "com.github.apple.coremltools.source" : "torch==2.5.1", "com.github.apple.coremltools.version" : "8.0" }, "inputSchema" : [ { "hasShapeFlexibility" : "0", "isOptional" : "0", "dataType" : "Float16", "formattedType" : "MultiArray (Float16 1 × 80 × 1 × 3000)", "shortDescription" : "", "shape" : "[1, 80, 1, 3000]", "name" : "melspectrogram_features", "type" : "MultiArray" } ], "generatedClassName" : "AudioEncoderStateful", "method" : "predict" } ]