| [ |
| { |
| "metadataOutputVersion" : "3.0", |
| "shortDescription" : "Fluid Speech Pro by FluidVoice (fp16): encoder with fv_insert (15 s window)", |
| "outputSchema" : [ |
| { |
| "hasShapeFlexibility" : "0", |
| "isOptional" : "0", |
| "dataType" : "Float32", |
| "formattedType" : "MultiArray (Float32 1 × 1024 × 188)", |
| "shortDescription" : "", |
| "shape" : "[1, 1024, 188]", |
| "name" : "encoder", |
| "type" : "MultiArray" |
| }, |
| { |
| "hasShapeFlexibility" : "0", |
| "isOptional" : "0", |
| "dataType" : "Int32", |
| "formattedType" : "MultiArray (Int32 1)", |
| "shortDescription" : "", |
| "shape" : "[1]", |
| "name" : "encoder_length", |
| "type" : "MultiArray" |
| } |
| ], |
| "storagePrecision" : "Mixed (Float16, Int32)", |
| "modelParameters" : [ |
|
|
| ], |
| "author" : "FluidVoice", |
| "specificationVersion" : 9, |
| "license" : "CC-BY-4.0", |
| "mlProgramOperationTypeHistogram" : { |
| "Ios18.expandDims" : 5, |
| "Ios18.mul" : 99, |
| "Ios18.softmax" : 24, |
| "Ios18.matmul" : 72, |
| "Ios18.logicalNot" : 2, |
| "Ios18.floor" : 3, |
| "Split" : 24, |
| "Ios18.sigmoid" : 24, |
| "Select" : 72, |
| "Tile" : 1, |
| "Ios18.add" : 175, |
| "Ios18.layerNorm" : 121, |
| "Ios18.reshape" : 145, |
| "Pad" : 48, |
| "Ios18.linear" : 195, |
| "Ios18.conv" : 78, |
| "Ios18.relu" : 3, |
| "Ios18.gelu" : 1, |
| "Ios18.logicalAnd" : 2, |
| "Ios18.silu" : 72, |
| "Ios18.transpose" : 174, |
| "Ios18.cast" : 4, |
| "Ios18.less" : 1, |
| "Ios18.sliceByIndex" : 48 |
| }, |
| "computePrecision" : "Mixed (Float16, Float32, Int32)", |
| "stateSchema" : [ |
|
|
| ], |
| "isUpdatable" : "0", |
| "availability" : { |
| "macOS" : "15.0", |
| "tvOS" : "18.0", |
| "visionOS" : "2.0", |
| "watchOS" : "11.0", |
| "iOS" : "18.0", |
| "macCatalyst" : "18.0" |
| }, |
| "modelType" : { |
| "name" : "MLModelType_mlProgram" |
| }, |
| "inputSchema" : [ |
| { |
| "hasShapeFlexibility" : "0", |
| "isOptional" : "0", |
| "dataType" : "Float32", |
| "formattedType" : "MultiArray (Float32 1 × 128 × 1501)", |
| "shortDescription" : "", |
| "shape" : "[1, 128, 1501]", |
| "name" : "mel", |
| "type" : "MultiArray" |
| }, |
| { |
| "hasShapeFlexibility" : "0", |
| "isOptional" : "0", |
| "dataType" : "Int32", |
| "formattedType" : "MultiArray (Int32 1)", |
| "shortDescription" : "", |
| "shape" : "[1]", |
| "name" : "mel_length", |
| "type" : "MultiArray" |
| } |
| ], |
| "userDefinedMetadata" : { |
| "com.github.apple.coremltools.conversion_date" : "2026-11-02", |
| "com.github.apple.coremltools.source_dialect" : "TorchScript", |
| "com.github.apple.coremltools.source" : "torch==2.7.0", |
| "com.github.apple.coremltools.version" : "9.0", |
| "fluidspeech.source" : "FluidVoice-Labs\/fluid-speech-pro@53f490b9292c5943847f635b7abfc3f123018964" |
| }, |
| "generatedClassName" : "Encoder", |
| "method" : "predict" |
| } |
| ] |
