| [ | |
| { | |
| "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" | |
| } | |
| ], | |
| "modelParameters" : [ | |
| ], | |
| "specificationVersion" : 7, | |
| "mlProgramOperationTypeHistogram" : { | |
| "Concat" : 28, | |
| "Ios16.rsqrt" : 9, | |
| "Ios16.mul" : 114, | |
| "SliceByIndex" : 168, | |
| "Ios16.sub" : 9, | |
| "Transpose" : 4, | |
| "Ios16.einsum" : 192, | |
| "Ios16.conv" : 26, | |
| "Ios16.add" : 18, | |
| "Ios16.reduceMean" : 18, | |
| "Ios16.softmax" : 96, | |
| "Ios16.gelu" : 6, | |
| "Ios16.batchNorm" : 9 | |
| }, | |
| "computePrecision" : "Mixed (Float16, Int32)", | |
| "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" | |
| }, | |
| "userDefinedMetadata" : { | |
| "com.github.apple.coremltools.source_dialect" : "TorchScript", | |
| "com.github.apple.coremltools.version" : "7.1", | |
| "com.github.apple.coremltools.source" : "torch==2.2.1" | |
| }, | |
| "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" : "AudioEncoder", | |
| "method" : "predict" | |
| } | |
| ] |