Upload folder using huggingface_hub

.gitignore

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

README.md

@@ -0,0 +1,24 @@
+
+---
+pretty_name: "WhisperKit"
+viewer: false
+library_name: whisperkit
+tags:
+- whisper
+- whisperkit
+- coreml
+- asr
+- quantized
+- automatic-speech-recognition
+---
+# WhisperKit
+
+WhisperKit is an on-device speech recognition framework for Apple Silicon:
+https://github.com/argmaxinc/WhisperKit
+
+For performance and accuracy benchmarks on real devices, please see:
+https://huggingface.co/spaces/argmaxinc/whisperkit-benchmarks
+
+WhisperKit Pro is the commercial upgraded version of WhisperKit. If you would like to learn more about upgrading to Pro, please reach out to [email protected] or fill the form [here](https://huggingface.co/argmaxinc/whisperkit-pro).
+
+

config.json

@@ -0,0 +1,219 @@
+{
+    "name": "whisperkit-coreml",
+    "version": "0.4",
+    "device_support": [
+        {
+            "chips": "A12, A13, S9, S10",
+            "identifiers": [
+                "iPhone11",
+                "iPhone12",
+                "iPad12,1",
+                "iPad12,2",
+                "Watch7",
+                "Watch8"
+            ],
+            "models": {
+                "default": "openai_whisper-tiny",
+                "supported": [
+                    "openai_whisper-tiny",
+                    "openai_whisper-tiny.en",
+                    "openai_whisper-base",
+                    "openai_whisper-base.en"
+                ]
+            }
+        },
+        {
+            "chips": "A16, A17 Pro, A18, A19",
+            "identifiers": [
+                "iPhone15",
+                "iPhone16",
+                "iPhone17",
+                "iPhone18",
+                "iPad15,7",
+                "iPad15,8",
+                "iPad16,1",
+                "iPad16,2"
+            ],
+            "models": {
+                "default": "openai_whisper-base",
+                "supported": [
+                    "openai_whisper-tiny",
+                    "openai_whisper-tiny.en",
+                    "openai_whisper-base",
+                    "openai_whisper-base.en",
+                    "openai_whisper-small",
+                    "openai_whisper-small.en",
+                    "openai_whisper-large-v2_949MB",
+                    "openai_whisper-large-v2_turbo_955MB",
+                    "openai_whisper-large-v3_947MB",
+                    "openai_whisper-large-v3_turbo_954MB",
+                    "distil-whisper_distil-large-v3_594MB",
+                    "distil-whisper_distil-large-v3_turbo_600MB",
+                    "openai_whisper-large-v3-v20240930_626MB",
+                    "openai_whisper-large-v3-v20240930_turbo_632MB"
+                ]
+            }
+        },
+        {
+            "chips": "M1",
+            "identifiers": [
+                "MacBookPro17,1",
+                "MacBookPro18,1",
+                "MacBookPro18,2",
+                "MacBookPro18,3",
+                "MacBookPro18,4",
+                "MacBookAir10,1",
+                "Macmini9,1",
+                "iMac21,1",
+                "iMac21,2",
+                "Mac13",
+                "iPad13,4",
+                "iPad13,5",
+                "iPad13,6",
+                "iPad13,7",
+                "iPad13,8",
+                "iPad13,9",
+                "iPad13,10",
+                "iPad13,11",
+                "iPad13,16",
+                "iPad13,17"
+            ],
+            "models": {
+                "default": "openai_whisper-large-v3-v20240930_626MB",
+                "supported": [
+                    "openai_whisper-tiny",
+                    "openai_whisper-tiny.en",
+                    "openai_whisper-base",
+                    "openai_whisper-base.en",
+                    "openai_whisper-small",
+                    "openai_whisper-small.en",
+                    "openai_whisper-large-v2",
+                    "openai_whisper-large-v2_949MB",
+                    "openai_whisper-large-v3",
+                    "openai_whisper-large-v3_947MB",
+                    "distil-whisper_distil-large-v3",
+                    "distil-whisper_distil-large-v3_594MB",
+                    "openai_whisper-large-v3-v20240930_626MB"
+                ]
+            }
+        },
+        {
+            "chips": "M2, M3, M4",
+            "identifiers": [
+                "Mac14",
+                "Mac15",
+                "Mac16",
+                "iPad14,3",
+                "iPad14,4",
+                "iPad14,5",
+                "iPad14,6",
+                "iPad14,8",
+                "iPad14,9",
+                "iPad14,10",
+                "iPad14,11",
+                "iPad15",
+                "iPad16"
+            ],
+            "models": {
+                "default": "openai_whisper-large-v3-v20240930",
+                "supported": [
+                    "openai_whisper-tiny",
+                    "openai_whisper-tiny.en",
+                    "openai_whisper-base",
+                    "openai_whisper-base.en",
+                    "openai_whisper-small",
+                    "openai_whisper-small.en",
+                    "openai_whisper-large-v2",
+                    "openai_whisper-large-v2_949MB",
+                    "openai_whisper-large-v2_turbo",
+                    "openai_whisper-large-v2_turbo_955MB",
+                    "openai_whisper-large-v3",
+                    "openai_whisper-large-v3_947MB",
+                    "openai_whisper-large-v3_turbo",
+                    "openai_whisper-large-v3_turbo_954MB",
+                    "distil-whisper_distil-large-v3",
+                    "distil-whisper_distil-large-v3_594MB",
+                    "distil-whisper_distil-large-v3_turbo",
+                    "distil-whisper_distil-large-v3_turbo_600MB",
+                    "openai_whisper-large-v3-v20240930",
+                    "openai_whisper-large-v3-v20240930_turbo",
+                    "openai_whisper-large-v3-v20240930_626MB",
+                    "openai_whisper-large-v3-v20240930_turbo_632MB"
+                ]
+            }
+        },
+        {
+            "chips": "A14",
+            "identifiers": [
+                "iPhone13",
+                "iPad13,1",
+                "iPad13,2",
+                "iPad13,18",
+                "iPad13,19"
+            ],
+            "models": {
+                "default": "openai_whisper-base",
+                "supported": [
+                    "openai_whisper-tiny",
+                    "openai_whisper-tiny.en",
+                    "openai_whisper-base",
+                    "openai_whisper-base.en",
+                    "openai_whisper-small",
+                    "openai_whisper-small.en"
+                ]
+            }
+        },
+        {
+            "chips": "A15",
+            "identifiers": [
+                "iPhone14",
+                "iPad14,1",
+                "iPad14,2"
+            ],
+            "models": {
+                "default": "openai_whisper-base",
+                "supported": [
+                    "openai_whisper-tiny",
+                    "openai_whisper-tiny.en",
+                    "openai_whisper-base",
+                    "openai_whisper-base.en",
+                    "openai_whisper-small",
+                    "openai_whisper-small.en",
+                    "openai_whisper-large-v2_949MB",
+                    "openai_whisper-large-v2_turbo_955MB",
+                    "openai_whisper-large-v3_947MB",
+                    "openai_whisper-large-v3_turbo_954MB",
+                    "distil-whisper_distil-large-v3_594MB",
+                    "distil-whisper_distil-large-v3_turbo_600MB",
+                    "openai_whisper-large-v3-v20240930_626MB",
+                    "openai_whisper-large-v3-v20240930_turbo_632MB"
+                ]
+            }
+        }
+    ],
+    "model_checksums": {
+        "distil-whisper_distil-large-v3": "9cd8271143b919402ae776c30b479565",
+        "distil-whisper_distil-large-v3_594MB": "ca532f45ddbf8a3d241132cc5cf41639",
+        "distil-whisper_distil-large-v3_turbo": "b8638452c6568dfe33a33bfcc2bc6aca",
+        "distil-whisper_distil-large-v3_turbo_600MB": "81746b4b1afbbb01a8ae9ea452460d88",
+        "openai_whisper-base.en": "fbcfd586f15e2952251b1d3257f18471",
+        "openai_whisper-base": "36e60501ad0f01c1a5719e83a1f63f20",
+        "openai_whisper-large-v2": "21b86c07318aeeef54598f15b7903979",
+        "openai_whisper-large-v2_949MB": "71bad4e1566749d1060eda42308d9fb4",
+        "openai_whisper-large-v2_turbo": "7734959b6550e7b5c2d732bf2b7acd23",
+        "openai_whisper-large-v2_turbo_955MB": "cb6411862a48ec75325572081f01e5b5",
+        "openai_whisper-large-v3-v20240930": "17ebd78ff7edfa59001b554e9cc4c021",
+        "openai_whisper-large-v3-v20240930_547MB": "c945dad68449ac3c78ecb2d561ac189d",
+        "openai_whisper-large-v3-v20240930_626MB": "578fe5a07f4eb7e4187c920bca571aa5",
+        "openai_whisper-large-v3-v20240930_turbo": "dfbf09ab741af1d5400ddbd07bb37dad",
+        "openai_whisper-large-v3-v20240930_turbo_632MB": "33954440dbd785ca1828afe25514f5a5",
+        "openai_whisper-large-v3": "a6f24dc72785722e9cea89e227856dfe",
+        "openai_whisper-large-v3_947MB": "ef6b0e9622a046ce2361b4c72307877f",
+        "openai_whisper-large-v3_turbo": "c550fbdea70c5784d322c0a427f8b5cd",
+        "openai_whisper-large-v3_turbo_954MB": "e639c4bb98d905064ef5dd38757dd9d1",
+        "openai_whisper-small.en": "38efe6a00706bbdb995795c67a836e5e",
+        "openai_whisper-small": "f1d21adb950bc9be5d5343bcdeccd23b",
+        "openai_whisper-tiny.en": "e1183fd55448923b1ce43a2da67aa21f",
+        "openai_whisper-tiny": "7147518a3d68ddbea0691e04cfffa4ff"
+    }
+}

distil-whisper_distil-large-v3/AudioEncoder.mlmodelc/analytics/coremldata.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:195e35e9fbaa218cd59eca69988a2165245e5bfccc4a5d9776847e8955d1625a
+size 243

distil-whisper_distil-large-v3/AudioEncoder.mlmodelc/coremldata.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:273d6cd004f95763e9d03e5d36622f11038819a81b9eafed64b1d95444e04f62
+size 348

distil-whisper_distil-large-v3/AudioEncoder.mlmodelc/metadata.json

@@ -0,0 +1,67 @@
+[
+  {
+    "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" : 130,
+      "Ios16.mul" : 162,
+      "Ios16.rsqrt" : 65,
+      "Ios16.sub" : 65,
+      "Ios16.batchNorm" : 65,
+      "Ios16.gelu" : 34,
+      "Ios16.reduceMean" : 130,
+      "Ios16.matmul" : 64,
+      "Ios16.reshape" : 128,
+      "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.source" : "torch==2.2.1",
+      "com.github.apple.coremltools.version" : "7.1"
+    },
+    "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"
+  }
+]

distil-whisper_distil-large-v3/AudioEncoder.mlmodelc/model.mlmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bf0987c0d8c3fe180877b12ba5b4ac1d890e011f2103926e53186089feced575
+size 408667

distil-whisper_distil-large-v3/AudioEncoder.mlmodelc/weights/weight.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8406a888fff74171e94d937de18b722373f083232782132ffbe5cd0ecc0f41db
+size 1273974400

distil-whisper_distil-large-v3/MelSpectrogram.mlmodelc/analytics/coremldata.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:091a361134891f94e613562771beea0d93a9aefbc6984ba86c60f856e07a508f
+size 243

distil-whisper_distil-large-v3/MelSpectrogram.mlmodelc/coremldata.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f3c5778c86d6fbc6a9817a56dbcac05a946a4d95c77f6db8355572f3be9e9a68
+size 329

distil-whisper_distil-large-v3/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" : 7,
+    "mlProgramOperationTypeHistogram" : {
+      "Pad" : 1,
+      "Ios16.mul" : 2,
+      "SliceByIndex" : 1,
+      "Ios16.sub" : 1,
+      "Ios16.log" : 1,
+      "Ios16.conv" : 2,
+      "Ios16.add" : 3,
+      "Ios16.square" : 2,
+      "Ios16.matmul" : 1,
+      "Squeeze" : 2,
+      "Ios16.maximum" : 1,
+      "ExpandDims" : 4,
+      "Ios16.reduceMax" : 1,
+      "Identity" : 1,
+      "Ios16.reshape" : 2
+    },
+    "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==2.2.1",
+      "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"
+  }
+]

distil-whisper_distil-large-v3/MelSpectrogram.mlmodelc/model.mil

@@ -0,0 +1,66 @@
+program(1.0)
+[buildInfo = dict<tensor<string, []>, tensor<string, []>>({{"coremlc-component-MIL", "5.33.5"}, {"coremlc-version", "1877.40.3"}, {"coremltools-component-torch", "2.2.1"}, {"coremltools-source-dialect", "TorchScript"}, {"coremltools-version", "7.1"}})]
+{
+    func main<ios16>(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<fp16, []> log_0_epsilon_0_to_fp16 = const()[name = tensor<string, []>("log_0_epsilon_0_to_fp16"), val = tensor<fp16, []>(0x0p+0)];
+            tensor<fp16, [128, 3000]> log_0_cast_fp16 = log(epsilon = log_0_epsilon_0_to_fp16, 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);
+}

distil-whisper_distil-large-v3/MelSpectrogram.mlmodelc/weights/weight.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f0c972cf2fd8de717c5129818259ff2a79ba41552b51a024c8d89eaf6fe4496
+size 373376

distil-whisper_distil-large-v3/TextDecoder.mlmodelc/analytics/coremldata.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a7a330a17952d9875fbb276af9f556979618bbe6b3087e22efe96a92f51efac2
+size 243

distil-whisper_distil-large-v3/TextDecoder.mlmodelc/coremldata.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a7372960953afc4893051757be8b0f78b1ed2d582ef5822b806adde589e097a8
+size 633

distil-whisper_distil-large-v3/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 × 2560 × 1 × 1)",
+        "shortDescription" : "",
+        "shape" : "[1, 2560, 1, 1]",
+        "name" : "key_cache_updates",
+        "type" : "MultiArray"
+      },
+      {
+        "hasShapeFlexibility" : "0",
+        "isOptional" : "0",
+        "dataType" : "Float16",
+        "formattedType" : "MultiArray (Float16 1 × 2560 × 1 × 1)",
+        "shortDescription" : "",
+        "shape" : "[1, 2560, 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,
+      "Ios16.rsqrt" : 7,
+      "Ios16.mul" : 26,
+      "Squeeze" : 1,
+      "SliceByIndex" : 40,
+      "Ios16.sub" : 8,
+      "Transpose" : 1,
+      "Ios16.conv" : 20,
+      "Ios16.add" : 20,
+      "Ios16.linear" : 1,
+      "Ios16.matmul" : 8,
+      "Ios16.gelu" : 2,
+      "Ios16.reduceMean" : 15,
+      "ExpandDims" : 6,
+      "Ios16.batchNorm" : 7,
+      "Ios16.gather" : 2,
+      "Ios16.reshape" : 16,
+      "Ios16.softmax" : 4
+    },
+    "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==2.2.1",
+      "com.github.apple.coremltools.version" : "7.1"
+    },
+    "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 × 2560 × 1 × 448)",
+        "shortDescription" : "",
+        "shape" : "[1, 2560, 1, 448]",
+        "name" : "key_cache",
+        "type" : "MultiArray"
+      },
+      {
+        "hasShapeFlexibility" : "0",
+        "isOptional" : "0",
+        "dataType" : "Float16",
+        "formattedType" : "MultiArray (Float16 1 × 2560 × 1 × 448)",
+        "shortDescription" : "",
+        "shape" : "[1, 2560, 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"
+  }
+]

distil-whisper_distil-large-v3/TextDecoder.mlmodelc/model.mlmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1a7ffcd457ab6ed7bc5e6f8dc699807c7605ffbba71940662332d9b7fe599503
+size 83662

distil-whisper_distil-large-v3/TextDecoder.mlmodelc/weights/weight.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0a38bc3b78a21d5bd912bd8ce72762456f8cdcfdfec5ce3767d678584e6e8d6c
+size 238986036

distil-whisper_distil-large-v3/config.json

@@ -0,0 +1 @@
+{"_name_or_path": "./distil-large-v3", "activation_dropout": 0.0, "activation_function": "gelu", "apply_spec_augment": false, "architectures": ["WhisperForConditionalGeneration"], "attention_dropout": 0.0, "begin_suppress_tokens": [220, 50257], "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": 2, "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_length": 448, "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": 50256, "scale_embedding": false, "torch_dtype": "float16", "transformers_version": "4.38.0.dev0", "use_cache": true, "use_weighted_layer_sum": false, "vocab_size": 51866}

distil-whisper_distil-large-v3/generation_config.json

@@ -0,0 +1 @@
+{"alignment_heads": [[7, 0], [10, 17], [12, 18], [13, 12], [16, 1], [17, 14], [19, 11], [21, 4], [24, 1], [25, 6]], "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}, "language": "<|en|>", "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": "transcribe", "task_to_id": {"transcribe": 50360, "translate": 50359}, "transformers_version": "4.38.0.dev0"}

distil-whisper_distil-large-v3_594MB/AudioEncoder.mlmodelc/analytics/coremldata.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:79120bd7c421c5c29ef5701ec2df116ef1b663951a3a0ac183147e015d35f2ac
+size 243

distil-whisper_distil-large-v3_594MB/AudioEncoder.mlmodelc/coremldata.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ed121e756d637301566cc86a4787e2adece71c438774ed1e5652a76d4f9ed62b
+size 348

distil-whisper_distil-large-v3_594MB/AudioEncoder.mlmodelc/metadata.json

@@ -0,0 +1,68 @@
+[
+  {
+    "metadataOutputVersion" : "3.0",
+    "storagePrecision" : "Mixed (Float16, Palettized (4 bits))",
+    "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" : 8,
+    "mlProgramOperationTypeHistogram" : {
+      "Ios17.mul" : 162,
+      "Ios17.rsqrt" : 65,
+      "Ios16.constexprLutToDense" : 192,
+      "Ios17.conv" : 578,
+      "Ios17.matmul" : 64,
+      "Ios17.sub" : 65,
+      "Ios17.add" : 322,
+      "Ios16.reduceMean" : 130,
+      "Ios17.batchNorm" : 65,
+      "Ios16.softmax" : 32,
+      "Ios16.gelu" : 34,
+      "Ios17.reshape" : 128
+    },
+    "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==2.2.0",
+      "com.github.apple.coremltools.version" : "7.1"
+    },
+    "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_mixedBitPalettized_4_0_bit",
+    "method" : "predict"
+  }
+]

distil-whisper_distil-large-v3_594MB/AudioEncoder.mlmodelc/weights/weight.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8f3ac62177a84bba905517f2546f559fff48c8b27eb57139f8dd7b11a283bc4d
+size 353908416

distil-whisper_distil-large-v3_594MB/MelSpectrogram.mlmodelc/analytics/coremldata.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:091a361134891f94e613562771beea0d93a9aefbc6984ba86c60f856e07a508f
+size 243

distil-whisper_distil-large-v3_594MB/MelSpectrogram.mlmodelc/coremldata.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f3c5778c86d6fbc6a9817a56dbcac05a946a4d95c77f6db8355572f3be9e9a68
+size 329

distil-whisper_distil-large-v3_594MB/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" : 7,
+    "mlProgramOperationTypeHistogram" : {
+      "Pad" : 1,
+      "Ios16.mul" : 2,
+      "SliceByIndex" : 1,
+      "Ios16.sub" : 1,
+      "Ios16.log" : 1,
+      "Ios16.conv" : 2,
+      "Ios16.add" : 3,
+      "Ios16.square" : 2,
+      "Ios16.matmul" : 1,
+      "Squeeze" : 2,
+      "Ios16.maximum" : 1,
+      "ExpandDims" : 4,
+      "Ios16.reduceMax" : 1,
+      "Identity" : 1,
+      "Ios16.reshape" : 2
+    },
+    "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==2.2.1",
+      "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"
+  }
+]

distil-whisper_distil-large-v3_594MB/MelSpectrogram.mlmodelc/model.mil

@@ -0,0 +1,66 @@
+program(1.0)
+[buildInfo = dict<tensor<string, []>, tensor<string, []>>({{"coremlc-component-MIL", "5.33.5"}, {"coremlc-version", "1877.40.3"}, {"coremltools-component-torch", "2.2.1"}, {"coremltools-source-dialect", "TorchScript"}, {"coremltools-version", "7.1"}})]
+{
+    func main<ios16>(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<fp16, []> log_0_epsilon_0_to_fp16 = const()[name = tensor<string, []>("log_0_epsilon_0_to_fp16"), val = tensor<fp16, []>(0x0p+0)];
+            tensor<fp16, [128, 3000]> log_0_cast_fp16 = log(epsilon = log_0_epsilon_0_to_fp16, 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);
+}

distil-whisper_distil-large-v3_594MB/MelSpectrogram.mlmodelc/weights/weight.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f0c972cf2fd8de717c5129818259ff2a79ba41552b51a024c8d89eaf6fe4496
+size 373376

distil-whisper_distil-large-v3_594MB/TextDecoder.mlmodelc/analytics/coremldata.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a7a330a17952d9875fbb276af9f556979618bbe6b3087e22efe96a92f51efac2
+size 243

distil-whisper_distil-large-v3_594MB/TextDecoder.mlmodelc/coremldata.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a7372960953afc4893051757be8b0f78b1ed2d582ef5822b806adde589e097a8
+size 633

distil-whisper_distil-large-v3_594MB/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 × 2560 × 1 × 1)",
+        "shortDescription" : "",
+        "shape" : "[1, 2560, 1, 1]",
+        "name" : "key_cache_updates",
+        "type" : "MultiArray"
+      },
+      {
+        "hasShapeFlexibility" : "0",
+        "isOptional" : "0",
+        "dataType" : "Float16",
+        "formattedType" : "MultiArray (Float16 1 × 2560 × 1 × 1)",
+        "shortDescription" : "",
+        "shape" : "[1, 2560, 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,
+      "Ios16.rsqrt" : 7,
+      "Ios16.mul" : 26,
+      "Squeeze" : 1,
+      "SliceByIndex" : 40,
+      "Ios16.sub" : 8,
+      "Transpose" : 1,
+      "Ios16.conv" : 20,
+      "Ios16.add" : 20,
+      "Ios16.linear" : 1,
+      "Ios16.matmul" : 8,
+      "Ios16.gelu" : 2,
+      "Ios16.reduceMean" : 15,
+      "ExpandDims" : 6,
+      "Ios16.batchNorm" : 7,
+      "Ios16.gather" : 2,
+      "Ios16.reshape" : 16,
+      "Ios16.softmax" : 4
+    },
+    "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==2.2.1",
+      "com.github.apple.coremltools.version" : "7.1"
+    },
+    "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 × 2560 × 1 × 448)",
+        "shortDescription" : "",
+        "shape" : "[1, 2560, 1, 448]",
+        "name" : "key_cache",
+        "type" : "MultiArray"
+      },
+      {
+        "hasShapeFlexibility" : "0",
+        "isOptional" : "0",
+        "dataType" : "Float16",
+        "formattedType" : "MultiArray (Float16 1 × 2560 × 1 × 448)",
+        "shortDescription" : "",
+        "shape" : "[1, 2560, 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"
+  }
+]

distil-whisper_distil-large-v3_594MB/TextDecoder.mlmodelc/weights/weight.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0a38bc3b78a21d5bd912bd8ce72762456f8cdcfdfec5ce3767d678584e6e8d6c
+size 238986036

distil-whisper_distil-large-v3_594MB/config.json

@@ -0,0 +1 @@
+{"_name_or_path": "./distil-large-v3", "activation_dropout": 0.0, "activation_function": "gelu", "apply_spec_augment": false, "architectures": ["WhisperForConditionalGeneration"], "attention_dropout": 0.0, "begin_suppress_tokens": [220, 50257], "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": 2, "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_length": 448, "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": 50256, "scale_embedding": false, "torch_dtype": "float16", "transformers_version": "4.38.0.dev0", "use_cache": true, "use_weighted_layer_sum": false, "vocab_size": 51866}

distil-whisper_distil-large-v3_594MB/generation_config.json

@@ -0,0 +1 @@
+{"alignment_heads": [[7, 0], [10, 17], [12, 18], [13, 12], [16, 1], [17, 14], [19, 11], [21, 4], [24, 1], [25, 6]], "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}, "language": "<|en|>", "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": "transcribe", "task_to_id": {"transcribe": 50360, "translate": 50359}, "transformers_version": "4.38.0.dev0"}

distil-whisper_distil-large-v3_turbo/AudioEncoder.mlmodelc/analytics/coremldata.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b68eb75cef8a4b3585423dca03d42d2f6ff34029421ad11752ea5be85c9bbf4c
+size 243

distil-whisper_distil-large-v3_turbo/AudioEncoder.mlmodelc/coremldata.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:273d6cd004f95763e9d03e5d36622f11038819a81b9eafed64b1d95444e04f62
+size 348

distil-whisper_distil-large-v3_turbo/AudioEncoder.mlmodelc/metadata.json

@@ -0,0 +1,69 @@
+[
+  {
+    "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" : {
+      "Concat" : 672,
+      "Ios16.rsqrt" : 65,
+      "Ios16.mul" : 2690,
+      "SliceByIndex" : 4480,
+      "Ios16.sub" : 65,
+      "Transpose" : 32,
+      "Ios16.einsum" : 5120,
+      "Ios16.conv" : 194,
+      "Ios16.add" : 130,
+      "Ios16.reduceMean" : 130,
+      "Ios16.softmax" : 2560,
+      "Ios16.gelu" : 34,
+      "Ios16.batchNorm" : 65
+    },
+    "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==2.2.1",
+      "com.github.apple.coremltools.version" : "7.1"
+    },
+    "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"
+  }
+]

distil-whisper_distil-large-v3_turbo/AudioEncoder.mlmodelc/model.mlmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bf0987c0d8c3fe180877b12ba5b4ac1d890e011f2103926e53186089feced575
+size 408667

distil-whisper_distil-large-v3_turbo/AudioEncoder.mlmodelc/weights/weight.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3a99a8d81ecb0ffa65d9e66dfc12da26a91c86cdd369db3d98aed777e8e15b68
+size 1273974400

distil-whisper_distil-large-v3_turbo/MelSpectrogram.mlmodelc/analytics/coremldata.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:091a361134891f94e613562771beea0d93a9aefbc6984ba86c60f856e07a508f
+size 243

distil-whisper_distil-large-v3_turbo/MelSpectrogram.mlmodelc/coremldata.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f3c5778c86d6fbc6a9817a56dbcac05a946a4d95c77f6db8355572f3be9e9a68
+size 329

distil-whisper_distil-large-v3_turbo/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" : 7,
+    "mlProgramOperationTypeHistogram" : {
+      "Pad" : 1,
+      "Ios16.mul" : 2,
+      "SliceByIndex" : 1,
+      "Ios16.sub" : 1,
+      "Ios16.log" : 1,
+      "Ios16.conv" : 2,
+      "Ios16.add" : 3,
+      "Ios16.square" : 2,
+      "Ios16.matmul" : 1,
+      "Squeeze" : 2,
+      "Ios16.maximum" : 1,
+      "ExpandDims" : 4,
+      "Ios16.reduceMax" : 1,
+      "Identity" : 1,
+      "Ios16.reshape" : 2
+    },
+    "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==2.2.1",
+      "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"
+  }
+]

distil-whisper_distil-large-v3_turbo/MelSpectrogram.mlmodelc/model.mil

@@ -0,0 +1,66 @@
+program(1.0)
+[buildInfo = dict<tensor<string, []>, tensor<string, []>>({{"coremlc-component-MIL", "5.33.5"}, {"coremlc-version", "1877.40.3"}, {"coremltools-component-torch", "2.2.1"}, {"coremltools-source-dialect", "TorchScript"}, {"coremltools-version", "7.1"}})]
+{
+    func main<ios16>(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<fp16, []> log_0_epsilon_0_to_fp16 = const()[name = tensor<string, []>("log_0_epsilon_0_to_fp16"), val = tensor<fp16, []>(0x0p+0)];
+            tensor<fp16, [128, 3000]> log_0_cast_fp16 = log(epsilon = log_0_epsilon_0_to_fp16, 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);
+}

distil-whisper_distil-large-v3_turbo/MelSpectrogram.mlmodelc/weights/weight.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a80bf139bea628069316903af7fb674a8d319233ab98f9d74b1c840915a4ef0
+size 373376