Support openai_whisper-large-v3-v20240930

.gitignore

@@ -0,0 +1 @@
+.DS_Store

openai_whisper-large-v3-v20240930/AudioEncoder.mlmodelc/analytics/coremldata.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:742c7bd747be2f2a27116a6ebe8ab4675babda518d2123f3efd774a890bc8304
+size 243

openai_whisper-large-v3-v20240930/AudioEncoder.mlmodelc/coremldata.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6345ee37d7bf7e45d4766046fa04822eb9e13bf144469541e9c1a3e0b911d2fe
+size 348

openai_whisper-large-v3-v20240930/AudioEncoder.mlmodelc/metadata.json

@@ -0,0 +1,65 @@
+[
+ {
+ "metadataOutputVersion" : "3.0",
+ "storagePrecision" : "Float16",
+ "outputSchema" : [
+ {
+ "hasShapeFlexibility" : "0",
+ "isOptional" : "0",
+ "dataType" : "Float16",
+ "formattedType" : "MultiArray (Float16 1 × 1280 × 1 × 1500)",
+ "shortDescription" : "",
+ "shape" : "[1, 1280, 1, 1500]",
+ "name" : "encoder_output_embeds",
+ "type" : "MultiArray"
+ }
+ ],
+ "modelParameters" : [
+
+ ],
+ "specificationVersion" : 7,
+ "mlProgramOperationTypeHistogram" : {
+ "Ios16.softmax" : 32,
+ "Ios16.add" : 65,
+ "Ios16.mul" : 32,
+ "Ios16.batchNorm" : 65,
+ "Ios16.gelu" : 34,
+ "Ios16.reshape" : 128,
+ "Ios16.matmul" : 64,
+ "Ios16.layerNorm" : 65,
+ "Ios16.conv" : 194
+ },
+ "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" : "8.0",
+ "com.github.apple.coremltools.source" : "torch==2.4.1"
+ },
+ "inputSchema" : [
+ {
+ "hasShapeFlexibility" : "0",
+ "isOptional" : "0",
+ "dataType" : "Float16",
+ "formattedType" : "MultiArray (Float16 1 × 128 × 1 × 3000)",
+ "shortDescription" : "",
+ "shape" : "[1, 128, 1, 3000]",
+ "name" : "melspectrogram_features",
+ "type" : "MultiArray"
+ }
+ ],
+ "generatedClassName" : "AudioEncoder",
+ "method" : "predict"
+ }
+]

openai_whisper-large-v3-v20240930/AudioEncoder.mlmodelc/weights/weight.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:98daf651a919978e28fe185daf55ce2f70085a8e59fa07fe8a4d08c87d368ae4
+size 1273974400

openai_whisper-large-v3-v20240930/MelSpectrogram.mlmodelc/analytics/coremldata.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7f478e6afa8fdee3711d97b00728f82ee4dcdf2912f7ff033ee0dfcf57bf28bb
+size 243

openai_whisper-large-v3-v20240930/MelSpectrogram.mlmodelc/coremldata.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a888718e98af679eee42db9e3609627472c32f77e4fdda28f3735960cbf526b3
+size 329

openai_whisper-large-v3-v20240930/MelSpectrogram.mlmodelc/metadata.json

@@ -0,0 +1,71 @@
+[
+ {
+ "metadataOutputVersion" : "3.0",
+ "storagePrecision" : "Float16",
+ "outputSchema" : [
+ {
+ "hasShapeFlexibility" : "0",
+ "isOptional" : "0",
+ "dataType" : "Float16",
+ "formattedType" : "MultiArray (Float16 1 × 128 × 1 × 3000)",
+ "shortDescription" : "",
+ "shape" : "[1, 128, 1, 3000]",
+ "name" : "melspectrogram_features",
+ "type" : "MultiArray"
+ }
+ ],
+ "modelParameters" : [
+
+ ],
+ "specificationVersion" : 8,
+ "mlProgramOperationTypeHistogram" : {
+ "Ios17.mul" : 2,
+ "Ios17.square" : 2,
+ "Ios17.sub" : 1,
+ "Ios17.matmul" : 1,
+ "Ios17.conv" : 2,
+ "Ios17.log" : 1,
+ "Ios17.add" : 3,
+ "Ios17.sliceByIndex" : 1,
+ "Ios17.expandDims" : 4,
+ "Ios17.maximum" : 1,
+ "Ios16.reduceMax" : 1,
+ "Ios17.squeeze" : 2,
+ "Ios17.reshape" : 2,
+ "Identity" : 1,
+ "Pad" : 1
+ },
+ "computePrecision" : "Mixed (Float16, Float32, Int32)",
+ "isUpdatable" : "0",
+ "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==2.2.1",
+ "com.github.apple.coremltools.version" : "7.1"
+ },
+ "inputSchema" : [
+ {
+ "hasShapeFlexibility" : "0",
+ "isOptional" : "0",
+ "dataType" : "Float16",
+ "formattedType" : "MultiArray (Float16 480000)",
+ "shortDescription" : "",
+ "shape" : "[480000]",
+ "name" : "audio",
+ "type" : "MultiArray"
+ }
+ ],
+ "generatedClassName" : "MelSpectrogram",
+ "method" : "predict"
+ }
+]

openai_whisper-large-v3-v20240930/MelSpectrogram.mlmodelc/model.mil

@@ -0,0 +1,66 @@
+program(1.0)
+[buildInfo = dict<tensor<string, []>, tensor<string, []>>({{"coremlc-component-MIL", "3304.5.2"}, {"coremlc-version", "3304.6.2"}, {"coremltools-component-torch", "2.2.1"}, {"coremltools-source-dialect", "TorchScript"}, {"coremltools-version", "7.1"}})]
+{
+ func main<ios17>(tensor<fp16, [480000]> audio) {
+ tensor<int32, [3]> var_10 = const()[name = tensor<string, []>("op_10"), val = tensor<int32, [3]>([1, 1, 480000])];
+ tensor<fp16, [1, 1, 480000]> input_1_cast_fp16 = reshape(shape = var_10, x = audio)[name = tensor<string, []>("input_1_cast_fp16")];
+ tensor<int32, [6]> input_3_pad_0 = const()[name = tensor<string, []>("input_3_pad_0"), val = tensor<int32, [6]>([0, 0, 0, 0, 200, 200])];
+ tensor<string, []> input_3_mode_0 = const()[name = tensor<string, []>("input_3_mode_0"), val = tensor<string, []>("reflect")];
+ tensor<fp16, []> input_3_constant_val_0_to_fp16 = const()[name = tensor<string, []>("input_3_constant_val_0_to_fp16"), val = tensor<fp16, []>(0x0p+0)];
+ tensor<fp16, [1, 1, 480400]> input_3_cast_fp16 = pad(constant_val = input_3_constant_val_0_to_fp16, mode = input_3_mode_0, pad = input_3_pad_0, x = input_1_cast_fp16)[name = tensor<string, []>("input_3_cast_fp16")];
+ tensor<int32, [1]> var_22 = const()[name = tensor<string, []>("op_22"), val = tensor<int32, [1]>([480400])];
+ tensor<fp16, [480400]> input_cast_fp16 = reshape(shape = var_22, x = input_3_cast_fp16)[name = tensor<string, []>("input_cast_fp16")];
+ tensor<int32, [1]> expand_dims_0_axes_0 = const()[name = tensor<string, []>("expand_dims_0_axes_0"), val = tensor<int32, [1]>([0])];
+ tensor<fp16, [1, 480400]> expand_dims_0_cast_fp16 = expand_dims(axes = expand_dims_0_axes_0, x = input_cast_fp16)[name = tensor<string, []>("expand_dims_0_cast_fp16")];
+ tensor<int32, [1]> expand_dims_3 = const()[name = tensor<string, []>("expand_dims_3"), val = tensor<int32, [1]>([160])];
+ tensor<int32, [1]> expand_dims_4_axes_0 = const()[name = tensor<string, []>("expand_dims_4_axes_0"), val = tensor<int32, [1]>([1])];
+ tensor<fp16, [1, 1, 480400]> expand_dims_4_cast_fp16 = expand_dims(axes = expand_dims_4_axes_0, x = expand_dims_0_cast_fp16)[name = tensor<string, []>("expand_dims_4_cast_fp16")];
+ tensor<string, []> conv_0_pad_type_0 = const()[name = tensor<string, []>("conv_0_pad_type_0"), val = tensor<string, []>("valid")];
+ tensor<int32, [2]> conv_0_pad_0 = const()[name = tensor<string, []>("conv_0_pad_0"), val = tensor<int32, [2]>([0, 0])];
+ tensor<int32, [1]> conv_0_dilations_0 = const()[name = tensor<string, []>("conv_0_dilations_0"), val = tensor<int32, [1]>([1])];
+ tensor<int32, []> conv_0_groups_0 = const()[name = tensor<string, []>("conv_0_groups_0"), val = tensor<int32, []>(1)];
+ tensor<fp16, [201, 1, 400]> expand_dims_1_to_fp16 = const()[name = tensor<string, []>("expand_dims_1_to_fp16"), val = tensor<fp16, [201, 1, 400]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(64)))];
+ tensor<fp16, [1, 201, 3001]> conv_0_cast_fp16 = conv(dilations = conv_0_dilations_0, groups = conv_0_groups_0, pad = conv_0_pad_0, pad_type = conv_0_pad_type_0, strides = expand_dims_3, weight = expand_dims_1_to_fp16, x = expand_dims_4_cast_fp16)[name = tensor<string, []>("conv_0_cast_fp16")];
+ tensor<string, []> conv_1_pad_type_0 = const()[name = tensor<string, []>("conv_1_pad_type_0"), val = tensor<string, []>("valid")];
+ tensor<int32, [2]> conv_1_pad_0 = const()[name = tensor<string, []>("conv_1_pad_0"), val = tensor<int32, [2]>([0, 0])];
+ tensor<int32, [1]> conv_1_dilations_0 = const()[name = tensor<string, []>("conv_1_dilations_0"), val = tensor<int32, [1]>([1])];
+ tensor<int32, []> conv_1_groups_0 = const()[name = tensor<string, []>("conv_1_groups_0"), val = tensor<int32, []>(1)];
+ tensor<fp16, [201, 1, 400]> expand_dims_2_to_fp16 = const()[name = tensor<string, []>("expand_dims_2_to_fp16"), val = tensor<fp16, [201, 1, 400]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(160960)))];
+ tensor<fp16, [1, 201, 3001]> conv_1_cast_fp16 = conv(dilations = conv_1_dilations_0, groups = conv_1_groups_0, pad = conv_1_pad_0, pad_type = conv_1_pad_type_0, strides = expand_dims_3, weight = expand_dims_2_to_fp16, x = expand_dims_4_cast_fp16)[name = tensor<string, []>("conv_1_cast_fp16")];
+ tensor<int32, [1]> squeeze_0_axes_0 = const()[name = tensor<string, []>("squeeze_0_axes_0"), val = tensor<int32, [1]>([0])];
+ tensor<fp16, [201, 3001]> squeeze_0_cast_fp16 = squeeze(axes = squeeze_0_axes_0, x = conv_0_cast_fp16)[name = tensor<string, []>("squeeze_0_cast_fp16")];
+ tensor<int32, [1]> squeeze_1_axes_0 = const()[name = tensor<string, []>("squeeze_1_axes_0"), val = tensor<int32, [1]>([0])];
+ tensor<fp16, [201, 3001]> squeeze_1_cast_fp16 = squeeze(axes = squeeze_1_axes_0, x = conv_1_cast_fp16)[name = tensor<string, []>("squeeze_1_cast_fp16")];
+ tensor<fp16, [201, 3001]> square_0_cast_fp16 = square(x = squeeze_0_cast_fp16)[name = tensor<string, []>("square_0_cast_fp16")];
+ tensor<fp16, [201, 3001]> square_1_cast_fp16 = square(x = squeeze_1_cast_fp16)[name = tensor<string, []>("square_1_cast_fp16")];
+ tensor<fp16, [201, 3001]> add_1_cast_fp16 = add(x = square_0_cast_fp16, y = square_1_cast_fp16)[name = tensor<string, []>("add_1_cast_fp16")];
+ tensor<fp16, [201, 3001]> magnitudes_1_cast_fp16 = identity(x = add_1_cast_fp16)[name = tensor<string, []>("magnitudes_1_cast_fp16")];
+ tensor<int32, [2]> magnitudes_begin_0 = const()[name = tensor<string, []>("magnitudes_begin_0"), val = tensor<int32, [2]>([0, 0])];
+ tensor<int32, [2]> magnitudes_end_0 = const()[name = tensor<string, []>("magnitudes_end_0"), val = tensor<int32, [2]>([201, 3000])];
+ tensor<bool, [2]> magnitudes_end_mask_0 = const()[name = tensor<string, []>("magnitudes_end_mask_0"), val = tensor<bool, [2]>([true, false])];
+ tensor<fp16, [201, 3000]> magnitudes_cast_fp16 = slice_by_index(begin = magnitudes_begin_0, end = magnitudes_end_0, end_mask = magnitudes_end_mask_0, x = magnitudes_1_cast_fp16)[name = tensor<string, []>("magnitudes_cast_fp16")];
+ tensor<bool, []> mel_spec_1_transpose_x_0 = const()[name = tensor<string, []>("mel_spec_1_transpose_x_0"), val = tensor<bool, []>(false)];
+ tensor<bool, []> mel_spec_1_transpose_y_0 = const()[name = tensor<string, []>("mel_spec_1_transpose_y_0"), val = tensor<bool, []>(false)];
+ tensor<fp16, [128, 201]> mel_filters_to_fp16 = const()[name = tensor<string, []>("mel_filters_to_fp16"), val = tensor<fp16, [128, 201]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(321856)))];
+ tensor<fp16, [128, 3000]> mel_spec_1_cast_fp16 = matmul(transpose_x = mel_spec_1_transpose_x_0, transpose_y = mel_spec_1_transpose_y_0, x = mel_filters_to_fp16, y = magnitudes_cast_fp16)[name = tensor<string, []>("mel_spec_1_cast_fp16")];
+ tensor<fp16, []> var_41_to_fp16 = const()[name = tensor<string, []>("op_41_to_fp16"), val = tensor<fp16, []>(0x1p-24)];
+ tensor<fp16, [128, 3000]> mel_spec_cast_fp16 = add(x = mel_spec_1_cast_fp16, y = var_41_to_fp16)[name = tensor<string, []>("mel_spec_cast_fp16")];
+ tensor<fp32, []> log_0_epsilon_0 = const()[name = tensor<string, []>("log_0_epsilon_0"), val = tensor<fp32, []>(0x1p-149)];
+ tensor<fp16, [128, 3000]> log_0_cast_fp16 = log(epsilon = log_0_epsilon_0, x = mel_spec_cast_fp16)[name = tensor<string, []>("log_0_cast_fp16")];
+ tensor<fp16, []> mul_0_y_0_to_fp16 = const()[name = tensor<string, []>("mul_0_y_0_to_fp16"), val = tensor<fp16, []>(0x1.bccp-2)];
+ tensor<fp16, [128, 3000]> mul_0_cast_fp16 = mul(x = log_0_cast_fp16, y = mul_0_y_0_to_fp16)[name = tensor<string, []>("mul_0_cast_fp16")];
+ tensor<bool, []> var_44_keep_dims_0 = const()[name = tensor<string, []>("op_44_keep_dims_0"), val = tensor<bool, []>(false)];
+ tensor<fp16, []> var_44_cast_fp16 = reduce_max(keep_dims = var_44_keep_dims_0, x = mul_0_cast_fp16)[name = tensor<string, []>("op_44_cast_fp16")];
+ tensor<fp16, []> var_46_to_fp16 = const()[name = tensor<string, []>("op_46_to_fp16"), val = tensor<fp16, []>(0x1p+3)];
+ tensor<fp16, []> var_47_cast_fp16 = sub(x = var_44_cast_fp16, y = var_46_to_fp16)[name = tensor<string, []>("op_47_cast_fp16")];
+ tensor<fp16, [128, 3000]> log_spec_3_cast_fp16 = maximum(x = mul_0_cast_fp16, y = var_47_cast_fp16)[name = tensor<string, []>("log_spec_3_cast_fp16")];
+ tensor<fp16, []> var_50_to_fp16 = const()[name = tensor<string, []>("op_50_to_fp16"), val = tensor<fp16, []>(0x1p+2)];
+ tensor<fp16, [128, 3000]> var_51_cast_fp16 = add(x = log_spec_3_cast_fp16, y = var_50_to_fp16)[name = tensor<string, []>("op_51_cast_fp16")];
+ tensor<fp16, []> _inversed_log_spec_y_0_to_fp16 = const()[name = tensor<string, []>("_inversed_log_spec_y_0_to_fp16"), val = tensor<fp16, []>(0x1p-2)];
+ tensor<fp16, [128, 3000]> _inversed_log_spec_cast_fp16 = mul(x = var_51_cast_fp16, y = _inversed_log_spec_y_0_to_fp16)[name = tensor<string, []>("_inversed_log_spec_cast_fp16")];
+ tensor<int32, [1]> var_55_axes_0 = const()[name = tensor<string, []>("op_55_axes_0"), val = tensor<int32, [1]>([0])];
+ tensor<fp16, [1, 128, 3000]> var_55_cast_fp16 = expand_dims(axes = var_55_axes_0, x = _inversed_log_spec_cast_fp16)[name = tensor<string, []>("op_55_cast_fp16")];
+ tensor<int32, [1]> var_62_axes_0 = const()[name = tensor<string, []>("op_62_axes_0"), val = tensor<int32, [1]>([2])];
+ tensor<fp16, [1, 128, 1, 3000]> melspectrogram_features = expand_dims(axes = var_62_axes_0, x = var_55_cast_fp16)[name = tensor<string, []>("op_62_cast_fp16")];
+ } -> (melspectrogram_features);
+}

openai_whisper-large-v3-v20240930/MelSpectrogram.mlmodelc/weights/weight.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:81275398516781f9755514a5ab85db4687374dd611013625f3d4493588783968
+size 373376

openai_whisper-large-v3-v20240930/TextDecoder.mlmodelc/analytics/coremldata.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:931efc5180111dbab2542c200f10b326d760c65a42dcf6c5edef73d803b93255
+size 243

openai_whisper-large-v3-v20240930/TextDecoder.mlmodelc/coremldata.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3faabaf66930e66956d8291d0ff485fb382496e30a91a7185548b9b898ce90a9
+size 633

openai_whisper-large-v3-v20240930/TextDecoder.mlmodelc/metadata.json

@@ -0,0 +1,165 @@
+[
+ {
+ "metadataOutputVersion" : "3.0",
+ "storagePrecision" : "Float16",
+ "outputSchema" : [
+ {
+ "hasShapeFlexibility" : "0",
+ "isOptional" : "0",
+ "dataType" : "Float16",
+ "formattedType" : "MultiArray (Float16 1 × 1 × 51866)",
+ "shortDescription" : "",
+ "shape" : "[1, 1, 51866]",
+ "name" : "logits",
+ "type" : "MultiArray"
+ },
+ {
+ "hasShapeFlexibility" : "0",
+ "isOptional" : "0",
+ "dataType" : "Float16",
+ "formattedType" : "MultiArray (Float16 1 × 5120 × 1 × 1)",
+ "shortDescription" : "",
+ "shape" : "[1, 5120, 1, 1]",
+ "name" : "key_cache_updates",
+ "type" : "MultiArray"
+ },
+ {
+ "hasShapeFlexibility" : "0",
+ "isOptional" : "0",
+ "dataType" : "Float16",
+ "formattedType" : "MultiArray (Float16 1 × 5120 × 1 × 1)",
+ "shortDescription" : "",
+ "shape" : "[1, 5120, 1, 1]",
+ "name" : "value_cache_updates",
+ "type" : "MultiArray"
+ },
+ {
+ "hasShapeFlexibility" : "0",
+ "isOptional" : "0",
+ "dataType" : "Float16",
+ "formattedType" : "MultiArray (Float16 1 × 1500)",
+ "shortDescription" : "",
+ "shape" : "[1, 1500]",
+ "name" : "alignment_heads_weights",
+ "type" : "MultiArray"
+ }
+ ],
+ "modelParameters" : [
+
+ ],
+ "specificationVersion" : 7,
+ "mlProgramOperationTypeHistogram" : {
+ "Split" : 2,
+ "Concat" : 3,
+ "Squeeze" : 1,
+ "Ios16.mul" : 24,
+ "Ios16.layerNorm" : 13,
+ "SliceByIndex" : 12,
+ "Ios16.sub" : 1,
+ "Transpose" : 1,
+ "Ios16.conv" : 40,
+ "Ios16.add" : 25,
+ "Ios16.linear" : 1,
+ "Ios16.matmul" : 16,
+ "Ios16.gelu" : 4,
+ "Ios16.reduceMean" : 1,
+ "ExpandDims" : 6,
+ "Ios16.batchNorm" : 13,
+ "Ios16.gather" : 2,
+ "Ios16.reshape" : 32,
+ "Ios16.softmax" : 8
+ },
+ "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.source" : "torch==2.4.1",
+ "com.github.apple.coremltools.version" : "8.0"
+ },
+ "inputSchema" : [
+ {
+ "hasShapeFlexibility" : "0",
+ "isOptional" : "0",
+ "dataType" : "Int32",
+ "formattedType" : "MultiArray (Int32 1)",
+ "shortDescription" : "",
+ "shape" : "[1]",
+ "name" : "input_ids",
+ "type" : "MultiArray"
+ },
+ {
+ "hasShapeFlexibility" : "0",
+ "isOptional" : "0",
+ "dataType" : "Int32",
+ "formattedType" : "MultiArray (Int32 1)",
+ "shortDescription" : "",
+ "shape" : "[1]",
+ "name" : "cache_length",
+ "type" : "MultiArray"
+ },
+ {
+ "hasShapeFlexibility" : "0",
+ "isOptional" : "0",
+ "dataType" : "Float16",
+ "formattedType" : "MultiArray (Float16 1 × 5120 × 1 × 448)",
+ "shortDescription" : "",
+ "shape" : "[1, 5120, 1, 448]",
+ "name" : "key_cache",
+ "type" : "MultiArray"
+ },
+ {
+ "hasShapeFlexibility" : "0",
+ "isOptional" : "0",
+ "dataType" : "Float16",
+ "formattedType" : "MultiArray (Float16 1 × 5120 × 1 × 448)",
+ "shortDescription" : "",
+ "shape" : "[1, 5120, 1, 448]",
+ "name" : "value_cache",
+ "type" : "MultiArray"
+ },
+ {
+ "hasShapeFlexibility" : "0",
+ "isOptional" : "0",
+ "dataType" : "Float16",
+ "formattedType" : "MultiArray (Float16 1 × 448)",
+ "shortDescription" : "",
+ "shape" : "[1, 448]",
+ "name" : "kv_cache_update_mask",
+ "type" : "MultiArray"
+ },
+ {
+ "hasShapeFlexibility" : "0",
+ "isOptional" : "0",
+ "dataType" : "Float16",
+ "formattedType" : "MultiArray (Float16 1 × 1280 × 1 × 1500)",
+ "shortDescription" : "",
+ "shape" : "[1, 1280, 1, 1500]",
+ "name" : "encoder_output_embeds",
+ "type" : "MultiArray"
+ },
+ {
+ "hasShapeFlexibility" : "0",
+ "isOptional" : "0",
+ "dataType" : "Float16",
+ "formattedType" : "MultiArray (Float16 1 × 448)",
+ "shortDescription" : "",
+ "shape" : "[1, 448]",
+ "name" : "decoder_key_padding_mask",
+ "type" : "MultiArray"
+ }
+ ],
+ "generatedClassName" : "TextDecoder",
+ "method" : "predict"
+ }
+]

openai_whisper-large-v3-v20240930/TextDecoder.mlmodelc/weights/weight.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:47b2703aa37448e09cf2f06e45984fabd5ded4c34ba3400cec38a5294af39dc1
+size 343933748

openai_whisper-large-v3-v20240930/config.json

@@ -0,0 +1,49 @@
+{
+ "_name_or_path": "/raid/yoach/tmp_whisper_turbo",
+ "activation_dropout": 0.0,
+ "activation_function": "gelu",
+ "apply_spec_augment": false,
+ "architectures": [
+ "WhisperForConditionalGeneration"
+ ],
+ "attention_dropout": 0.0,
+ "begin_suppress_tokens": [
+ 220,
+ 50256
+ ],
+ "bos_token_id": 50257,
+ "classifier_proj_size": 256,
+ "d_model": 1280,
+ "decoder_attention_heads": 20,
+ "decoder_ffn_dim": 5120,
+ "decoder_layerdrop": 0.0,
+ "decoder_layers": 4,
+ "decoder_start_token_id": 50258,
+ "dropout": 0.0,
+ "encoder_attention_heads": 20,
+ "encoder_ffn_dim": 5120,
+ "encoder_layerdrop": 0.0,
+ "encoder_layers": 32,
+ "eos_token_id": 50257,
+ "init_std": 0.02,
+ "is_encoder_decoder": true,
+ "mask_feature_length": 10,
+ "mask_feature_min_masks": 0,
+ "mask_feature_prob": 0.0,
+ "mask_time_length": 10,
+ "mask_time_min_masks": 2,
+ "mask_time_prob": 0.05,
+ "max_source_positions": 1500,
+ "max_target_positions": 448,
+ "median_filter_width": 7,
+ "model_type": "whisper",
+ "num_hidden_layers": 32,
+ "num_mel_bins": 128,
+ "pad_token_id": 50257,
+ "scale_embedding": false,
+ "torch_dtype": "float16",
+ "transformers_version": "4.46.0.dev0",
+ "use_cache": true,
+ "use_weighted_layer_sum": false,
+ "vocab_size": 51866
+}

openai_whisper-large-v3-v20240930/generation_config.json

@@ -0,0 +1,249 @@
+{
+ "alignment_heads": [
+ [
+ 2,
+ 4
+ ],
+ [
+ 2,
+ 11
+ ],
+ [
+ 3,
+ 3
+ ],
+ [
+ 3,
+ 6
+ ],
+ [
+ 3,
+ 11
+ ],
+ [
+ 3,
+ 14
+ ]
+ ],
+ "begin_suppress_tokens": [
+ 220,
+ 50257
+ ],
+ "bos_token_id": 50257,
+ "decoder_start_token_id": 50258,
+ "eos_token_id": 50257,
+ "forced_decoder_ids": [
+ [
+ 1,
+ null
+ ],
+ [
+ 2,
+ 50360
+ ]
+ ],
+ "is_multilingual": true,
+ "lang_to_id": {
+ "<|af|>": 50327,
+ "<|am|>": 50334,
+ "<|ar|>": 50272,
+ "<|as|>": 50350,
+ "<|az|>": 50304,
+ "<|ba|>": 50355,
+ "<|be|>": 50330,
+ "<|bg|>": 50292,
+ "<|bn|>": 50302,
+ "<|bo|>": 50347,
+ "<|br|>": 50309,
+ "<|bs|>": 50315,
+ "<|ca|>": 50270,
+ "<|cs|>": 50283,
+ "<|cy|>": 50297,
+ "<|da|>": 50285,
+ "<|de|>": 50261,
+ "<|el|>": 50281,
+ "<|en|>": 50259,
+ "<|es|>": 50262,
+ "<|et|>": 50307,
+ "<|eu|>": 50310,
+ "<|fa|>": 50300,
+ "<|fi|>": 50277,
+ "<|fo|>": 50338,
+ "<|fr|>": 50265,
+ "<|gl|>": 50319,
+ "<|gu|>": 50333,
+ "<|haw|>": 50352,
+ "<|ha|>": 50354,
+ "<|he|>": 50279,
+ "<|hi|>": 50276,
+ "<|hr|>": 50291,
+ "<|ht|>": 50339,
+ "<|hu|>": 50286,
+ "<|hy|>": 50312,
+ "<|id|>": 50275,
+ "<|is|>": 50311,
+ "<|it|>": 50274,
+ "<|ja|>": 50266,
+ "<|jw|>": 50356,
+ "<|ka|>": 50329,
+ "<|kk|>": 50316,
+ "<|km|>": 50323,
+ "<|kn|>": 50306,
+ "<|ko|>": 50264,
+ "<|la|>": 50294,
+ "<|lb|>": 50345,
+ "<|ln|>": 50353,
+ "<|lo|>": 50336,
+ "<|lt|>": 50293,
+ "<|lv|>": 50301,
+ "<|mg|>": 50349,
+ "<|mi|>": 50295,
+ "<|mk|>": 50308,
+ "<|ml|>": 50296,
+ "<|mn|>": 50314,
+ "<|mr|>": 50320,
+ "<|ms|>": 50282,
+ "<|mt|>": 50343,
+ "<|my|>": 50346,
+ "<|ne|>": 50313,
+ "<|nl|>": 50271,
+ "<|nn|>": 50342,
+ "<|no|>": 50288,
+ "<|oc|>": 50328,
+ "<|pa|>": 50321,
+ "<|pl|>": 50269,
+ "<|ps|>": 50340,
+ "<|pt|>": 50267,
+ "<|ro|>": 50284,
+ "<|ru|>": 50263,
+ "<|sa|>": 50344,
+ "<|sd|>": 50332,
+ "<|si|>": 50322,
+ "<|sk|>": 50298,
+ "<|sl|>": 50305,
+ "<|sn|>": 50324,
+ "<|so|>": 50326,
+ "<|sq|>": 50317,
+ "<|sr|>": 50303,
+ "<|su|>": 50357,
+ "<|sv|>": 50273,
+ "<|sw|>": 50318,
+ "<|ta|>": 50287,
+ "<|te|>": 50299,
+ "<|tg|>": 50331,
+ "<|th|>": 50289,
+ "<|tk|>": 50341,
+ "<|tl|>": 50348,
+ "<|tr|>": 50268,
+ "<|tt|>": 50351,
+ "<|uk|>": 50280,
+ "<|ur|>": 50290,
+ "<|uz|>": 50337,
+ "<|vi|>": 50278,
+ "<|yi|>": 50335,
+ "<|yo|>": 50325,
+ "<|yue|>": 50358,
+ "<|zh|>": 50260
+ },
+ "max_initial_timestamp_index": 50,
+ "max_length": 448,
+ "no_timestamps_token_id": 50364,
+ "pad_token_id": 50257,
+ "prev_sot_token_id": 50362,
+ "return_timestamps": false,
+ "suppress_tokens": [
+ 1,
+ 2,
+ 7,
+ 8,
+ 9,
+ 10,
+ 14,
+ 25,
+ 26,
+ 27,
+ 28,
+ 29,
+ 31,
+ 58,
+ 59,
+ 60,
+ 61,
+ 62,
+ 63,
+ 90,
+ 91,
+ 92,
+ 93,
+ 359,
+ 503,
+ 522,
+ 542,
+ 873,
+ 893,
+ 902,
+ 918,
+ 922,
+ 931,
+ 1350,
+ 1853,
+ 1982,
+ 2460,
+ 2627,
+ 3246,
+ 3253,
+ 3268,
+ 3536,
+ 3846,
+ 3961,
+ 4183,
+ 4667,
+ 6585,
+ 6647,
+ 7273,
+ 9061,
+ 9383,
+ 10428,
+ 10929,
+ 11938,
+ 12033,
+ 12331,
+ 12562,
+ 13793,
+ 14157,
+ 14635,
+ 15265,
+ 15618,
+ 16553,
+ 16604,
+ 18362,
+ 18956,
+ 20075,
+ 21675,
+ 22520,
+ 26130,
+ 26161,
+ 26435,
+ 28279,
+ 29464,
+ 31650,
+ 32302,
+ 32470,
+ 36865,
+ 42863,
+ 47425,
+ 49870,
+ 50254,
+ 50258,
+ 50359,
+ 50360,
+ 50361,
+ 50362,
+ 50363
+ ],
+ "task_to_id": {
+ "transcribe": 50360,
+ "translate": 50359
+ },
+ "transformers_version": "4.46.0.dev0"
+}