brianmunarcajulis commited on
Commit
5ecbf42
1 Parent(s): 5b69704

Delete distil-whisper_distil-large-v3_turbo_600MB

Browse files
Files changed (22) hide show
  1. distil-whisper_distil-large-v3_turbo_600MB/AudioEncoder.mlmodelc/analytics/coremldata.bin +0 -3
  2. distil-whisper_distil-large-v3_turbo_600MB/AudioEncoder.mlmodelc/coremldata.bin +0 -3
  3. distil-whisper_distil-large-v3_turbo_600MB/AudioEncoder.mlmodelc/metadata.json +0 -70
  4. distil-whisper_distil-large-v3_turbo_600MB/AudioEncoder.mlmodelc/model.mil +0 -0
  5. distil-whisper_distil-large-v3_turbo_600MB/AudioEncoder.mlmodelc/weights/weight.bin +0 -3
  6. distil-whisper_distil-large-v3_turbo_600MB/MelSpectrogram.mlmodelc/analytics/coremldata.bin +0 -3
  7. distil-whisper_distil-large-v3_turbo_600MB/MelSpectrogram.mlmodelc/coremldata.bin +0 -3
  8. distil-whisper_distil-large-v3_turbo_600MB/MelSpectrogram.mlmodelc/metadata.json +0 -71
  9. distil-whisper_distil-large-v3_turbo_600MB/MelSpectrogram.mlmodelc/model.mil +0 -66
  10. distil-whisper_distil-large-v3_turbo_600MB/MelSpectrogram.mlmodelc/weights/weight.bin +0 -3
  11. distil-whisper_distil-large-v3_turbo_600MB/TextDecoder.mlmodelc/analytics/coremldata.bin +0 -3
  12. distil-whisper_distil-large-v3_turbo_600MB/TextDecoder.mlmodelc/coremldata.bin +0 -3
  13. distil-whisper_distil-large-v3_turbo_600MB/TextDecoder.mlmodelc/metadata.json +0 -165
  14. distil-whisper_distil-large-v3_turbo_600MB/TextDecoder.mlmodelc/model.mil +0 -0
  15. distil-whisper_distil-large-v3_turbo_600MB/TextDecoder.mlmodelc/weights/weight.bin +0 -3
  16. distil-whisper_distil-large-v3_turbo_600MB/TextDecoderContextPrefill.mlmodelc/analytics/coremldata.bin +0 -3
  17. distil-whisper_distil-large-v3_turbo_600MB/TextDecoderContextPrefill.mlmodelc/coremldata.bin +0 -3
  18. distil-whisper_distil-large-v3_turbo_600MB/TextDecoderContextPrefill.mlmodelc/metadata.json +0 -82
  19. distil-whisper_distil-large-v3_turbo_600MB/TextDecoderContextPrefill.mlmodelc/model.mil +0 -27
  20. distil-whisper_distil-large-v3_turbo_600MB/TextDecoderContextPrefill.mlmodelc/weights/weight.bin +0 -3
  21. distil-whisper_distil-large-v3_turbo_600MB/config.json +0 -1
  22. distil-whisper_distil-large-v3_turbo_600MB/generation_config.json +0 -1
distil-whisper_distil-large-v3_turbo_600MB/AudioEncoder.mlmodelc/analytics/coremldata.bin DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:94b7b86bea508311eb4f6f60fa2692bdd4ed566ecd54a80f463f22ece82d8a33
3
- size 243
 
 
 
 
distil-whisper_distil-large-v3_turbo_600MB/AudioEncoder.mlmodelc/coremldata.bin DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:ed121e756d637301566cc86a4787e2adece71c438774ed1e5652a76d4f9ed62b
3
- size 348
 
 
 
 
distil-whisper_distil-large-v3_turbo_600MB/AudioEncoder.mlmodelc/metadata.json DELETED
@@ -1,70 +0,0 @@
1
- [
2
- {
3
- "metadataOutputVersion" : "3.0",
4
- "storagePrecision" : "Mixed (Float16, Palettized (4 bits))",
5
- "outputSchema" : [
6
- {
7
- "hasShapeFlexibility" : "0",
8
- "isOptional" : "0",
9
- "dataType" : "Float16",
10
- "formattedType" : "MultiArray (Float16 1 × 1280 × 1 × 1500)",
11
- "shortDescription" : "",
12
- "shape" : "[1, 1280, 1, 1500]",
13
- "name" : "encoder_output_embeds",
14
- "type" : "MultiArray"
15
- }
16
- ],
17
- "modelParameters" : [
18
-
19
- ],
20
- "specificationVersion" : 8,
21
- "mlProgramOperationTypeHistogram" : {
22
- "Ios17.mul" : 2690,
23
- "Ios17.transpose" : 32,
24
- "Ios17.rsqrt" : 65,
25
- "Ios16.constexprLutToDense" : 192,
26
- "Ios17.conv" : 578,
27
- "Ios17.sub" : 65,
28
- "Ios16.einsum" : 5120,
29
- "Ios17.concat" : 672,
30
- "Ios17.add" : 322,
31
- "Ios16.reduceMean" : 130,
32
- "Ios17.batchNorm" : 65,
33
- "Ios17.sliceByIndex" : 4480,
34
- "Ios16.softmax" : 2560,
35
- "Ios16.gelu" : 34
36
- },
37
- "computePrecision" : "Mixed (Float16, Float32, Int32)",
38
- "isUpdatable" : "0",
39
- "availability" : {
40
- "macOS" : "14.0",
41
- "tvOS" : "17.0",
42
- "visionOS" : "1.0",
43
- "watchOS" : "10.0",
44
- "iOS" : "17.0",
45
- "macCatalyst" : "17.0"
46
- },
47
- "modelType" : {
48
- "name" : "MLModelType_mlProgram"
49
- },
50
- "userDefinedMetadata" : {
51
- "com.github.apple.coremltools.source_dialect" : "TorchScript",
52
- "com.github.apple.coremltools.source" : "torch==2.2.0",
53
- "com.github.apple.coremltools.version" : "7.1"
54
- },
55
- "inputSchema" : [
56
- {
57
- "hasShapeFlexibility" : "0",
58
- "isOptional" : "0",
59
- "dataType" : "Float16",
60
- "formattedType" : "MultiArray (Float16 1 × 128 × 1 × 3000)",
61
- "shortDescription" : "",
62
- "shape" : "[1, 128, 1, 3000]",
63
- "name" : "melspectrogram_features",
64
- "type" : "MultiArray"
65
- }
66
- ],
67
- "generatedClassName" : "AudioEncoder_mixedBitPalettized_4_0_bit",
68
- "method" : "predict"
69
- }
70
- ]
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
distil-whisper_distil-large-v3_turbo_600MB/AudioEncoder.mlmodelc/model.mil DELETED
The diff for this file is too large to render. See raw diff
 
distil-whisper_distil-large-v3_turbo_600MB/AudioEncoder.mlmodelc/weights/weight.bin DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:98410d88d5243d74e5e5210febfbb84dd7a9392793f0fb320b46ab8b92c52a15
3
- size 353908416
 
 
 
 
distil-whisper_distil-large-v3_turbo_600MB/MelSpectrogram.mlmodelc/analytics/coremldata.bin DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:091a361134891f94e613562771beea0d93a9aefbc6984ba86c60f856e07a508f
3
- size 243
 
 
 
 
distil-whisper_distil-large-v3_turbo_600MB/MelSpectrogram.mlmodelc/coremldata.bin DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:f3c5778c86d6fbc6a9817a56dbcac05a946a4d95c77f6db8355572f3be9e9a68
3
- size 329
 
 
 
 
distil-whisper_distil-large-v3_turbo_600MB/MelSpectrogram.mlmodelc/metadata.json DELETED
@@ -1,71 +0,0 @@
1
- [
2
- {
3
- "metadataOutputVersion" : "3.0",
4
- "storagePrecision" : "Float16",
5
- "outputSchema" : [
6
- {
7
- "hasShapeFlexibility" : "0",
8
- "isOptional" : "0",
9
- "dataType" : "Float16",
10
- "formattedType" : "MultiArray (Float16 1 × 128 × 1 × 3000)",
11
- "shortDescription" : "",
12
- "shape" : "[1, 128, 1, 3000]",
13
- "name" : "melspectrogram_features",
14
- "type" : "MultiArray"
15
- }
16
- ],
17
- "modelParameters" : [
18
-
19
- ],
20
- "specificationVersion" : 7,
21
- "mlProgramOperationTypeHistogram" : {
22
- "Pad" : 1,
23
- "Ios16.mul" : 2,
24
- "SliceByIndex" : 1,
25
- "Ios16.sub" : 1,
26
- "Ios16.log" : 1,
27
- "Ios16.conv" : 2,
28
- "Ios16.add" : 3,
29
- "Ios16.square" : 2,
30
- "Ios16.matmul" : 1,
31
- "Squeeze" : 2,
32
- "Ios16.maximum" : 1,
33
- "ExpandDims" : 4,
34
- "Ios16.reduceMax" : 1,
35
- "Identity" : 1,
36
- "Ios16.reshape" : 2
37
- },
38
- "computePrecision" : "Mixed (Float16, Int32)",
39
- "isUpdatable" : "0",
40
- "availability" : {
41
- "macOS" : "13.0",
42
- "tvOS" : "16.0",
43
- "visionOS" : "1.0",
44
- "watchOS" : "9.0",
45
- "iOS" : "16.0",
46
- "macCatalyst" : "16.0"
47
- },
48
- "modelType" : {
49
- "name" : "MLModelType_mlProgram"
50
- },
51
- "userDefinedMetadata" : {
52
- "com.github.apple.coremltools.source_dialect" : "TorchScript",
53
- "com.github.apple.coremltools.source" : "torch==2.2.1",
54
- "com.github.apple.coremltools.version" : "7.1"
55
- },
56
- "inputSchema" : [
57
- {
58
- "hasShapeFlexibility" : "0",
59
- "isOptional" : "0",
60
- "dataType" : "Float16",
61
- "formattedType" : "MultiArray (Float16 480000)",
62
- "shortDescription" : "",
63
- "shape" : "[480000]",
64
- "name" : "audio",
65
- "type" : "MultiArray"
66
- }
67
- ],
68
- "generatedClassName" : "MelSpectrogram",
69
- "method" : "predict"
70
- }
71
- ]
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
distil-whisper_distil-large-v3_turbo_600MB/MelSpectrogram.mlmodelc/model.mil DELETED
@@ -1,66 +0,0 @@
1
- program(1.0)
2
- [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"}})]
3
- {
4
- func main<ios16>(tensor<fp16, [480000]> audio) {
5
- tensor<int32, [3]> var_10 = const()[name = tensor<string, []>("op_10"), val = tensor<int32, [3]>([1, 1, 480000])];
6
- tensor<fp16, [1, 1, 480000]> input_1_cast_fp16 = reshape(shape = var_10, x = audio)[name = tensor<string, []>("input_1_cast_fp16")];
7
- 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])];
8
- tensor<string, []> input_3_mode_0 = const()[name = tensor<string, []>("input_3_mode_0"), val = tensor<string, []>("reflect")];
9
- 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)];
10
- 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")];
11
- tensor<int32, [1]> var_22 = const()[name = tensor<string, []>("op_22"), val = tensor<int32, [1]>([480400])];
12
- tensor<fp16, [480400]> input_cast_fp16 = reshape(shape = var_22, x = input_3_cast_fp16)[name = tensor<string, []>("input_cast_fp16")];
13
- tensor<int32, [1]> expand_dims_0_axes_0 = const()[name = tensor<string, []>("expand_dims_0_axes_0"), val = tensor<int32, [1]>([0])];
14
- 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")];
15
- tensor<int32, [1]> expand_dims_3 = const()[name = tensor<string, []>("expand_dims_3"), val = tensor<int32, [1]>([160])];
16
- tensor<int32, [1]> expand_dims_4_axes_0 = const()[name = tensor<string, []>("expand_dims_4_axes_0"), val = tensor<int32, [1]>([1])];
17
- 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")];
18
- tensor<string, []> conv_0_pad_type_0 = const()[name = tensor<string, []>("conv_0_pad_type_0"), val = tensor<string, []>("valid")];
19
- tensor<int32, [2]> conv_0_pad_0 = const()[name = tensor<string, []>("conv_0_pad_0"), val = tensor<int32, [2]>([0, 0])];
20
- tensor<int32, [1]> conv_0_dilations_0 = const()[name = tensor<string, []>("conv_0_dilations_0"), val = tensor<int32, [1]>([1])];
21
- tensor<int32, []> conv_0_groups_0 = const()[name = tensor<string, []>("conv_0_groups_0"), val = tensor<int32, []>(1)];
22
- 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)))];
23
- 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")];
24
- tensor<string, []> conv_1_pad_type_0 = const()[name = tensor<string, []>("conv_1_pad_type_0"), val = tensor<string, []>("valid")];
25
- tensor<int32, [2]> conv_1_pad_0 = const()[name = tensor<string, []>("conv_1_pad_0"), val = tensor<int32, [2]>([0, 0])];
26
- tensor<int32, [1]> conv_1_dilations_0 = const()[name = tensor<string, []>("conv_1_dilations_0"), val = tensor<int32, [1]>([1])];
27
- tensor<int32, []> conv_1_groups_0 = const()[name = tensor<string, []>("conv_1_groups_0"), val = tensor<int32, []>(1)];
28
- 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)))];
29
- 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")];
30
- tensor<int32, [1]> squeeze_0_axes_0 = const()[name = tensor<string, []>("squeeze_0_axes_0"), val = tensor<int32, [1]>([0])];
31
- 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")];
32
- tensor<int32, [1]> squeeze_1_axes_0 = const()[name = tensor<string, []>("squeeze_1_axes_0"), val = tensor<int32, [1]>([0])];
33
- 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")];
34
- tensor<fp16, [201, 3001]> square_0_cast_fp16 = square(x = squeeze_0_cast_fp16)[name = tensor<string, []>("square_0_cast_fp16")];
35
- tensor<fp16, [201, 3001]> square_1_cast_fp16 = square(x = squeeze_1_cast_fp16)[name = tensor<string, []>("square_1_cast_fp16")];
36
- 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")];
37
- tensor<fp16, [201, 3001]> magnitudes_1_cast_fp16 = identity(x = add_1_cast_fp16)[name = tensor<string, []>("magnitudes_1_cast_fp16")];
38
- tensor<int32, [2]> magnitudes_begin_0 = const()[name = tensor<string, []>("magnitudes_begin_0"), val = tensor<int32, [2]>([0, 0])];
39
- tensor<int32, [2]> magnitudes_end_0 = const()[name = tensor<string, []>("magnitudes_end_0"), val = tensor<int32, [2]>([201, 3000])];
40
- tensor<bool, [2]> magnitudes_end_mask_0 = const()[name = tensor<string, []>("magnitudes_end_mask_0"), val = tensor<bool, [2]>([true, false])];
41
- 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")];
42
- tensor<bool, []> mel_spec_1_transpose_x_0 = const()[name = tensor<string, []>("mel_spec_1_transpose_x_0"), val = tensor<bool, []>(false)];
43
- tensor<bool, []> mel_spec_1_transpose_y_0 = const()[name = tensor<string, []>("mel_spec_1_transpose_y_0"), val = tensor<bool, []>(false)];
44
- 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)))];
45
- 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")];
46
- tensor<fp16, []> var_41_to_fp16 = const()[name = tensor<string, []>("op_41_to_fp16"), val = tensor<fp16, []>(0x1p-24)];
47
- 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")];
48
- tensor<fp16, []> log_0_epsilon_0_to_fp16 = const()[name = tensor<string, []>("log_0_epsilon_0_to_fp16"), val = tensor<fp16, []>(0x0p+0)];
49
- 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")];
50
- tensor<fp16, []> mul_0_y_0_to_fp16 = const()[name = tensor<string, []>("mul_0_y_0_to_fp16"), val = tensor<fp16, []>(0x1.bccp-2)];
51
- 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")];
52
- tensor<bool, []> var_44_keep_dims_0 = const()[name = tensor<string, []>("op_44_keep_dims_0"), val = tensor<bool, []>(false)];
53
- 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")];
54
- tensor<fp16, []> var_46_to_fp16 = const()[name = tensor<string, []>("op_46_to_fp16"), val = tensor<fp16, []>(0x1p+3)];
55
- tensor<fp16, []> var_47_cast_fp16 = sub(x = var_44_cast_fp16, y = var_46_to_fp16)[name = tensor<string, []>("op_47_cast_fp16")];
56
- 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")];
57
- tensor<fp16, []> var_50_to_fp16 = const()[name = tensor<string, []>("op_50_to_fp16"), val = tensor<fp16, []>(0x1p+2)];
58
- 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")];
59
- 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)];
60
- 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")];
61
- tensor<int32, [1]> var_55_axes_0 = const()[name = tensor<string, []>("op_55_axes_0"), val = tensor<int32, [1]>([0])];
62
- 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")];
63
- tensor<int32, [1]> var_62_axes_0 = const()[name = tensor<string, []>("op_62_axes_0"), val = tensor<int32, [1]>([2])];
64
- 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")];
65
- } -> (melspectrogram_features);
66
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
distil-whisper_distil-large-v3_turbo_600MB/MelSpectrogram.mlmodelc/weights/weight.bin DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:7a80bf139bea628069316903af7fb674a8d319233ab98f9d74b1c840915a4ef0
3
- size 373376
 
 
 
 
distil-whisper_distil-large-v3_turbo_600MB/TextDecoder.mlmodelc/analytics/coremldata.bin DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:013f68d396d51f281bef4db5e0c3b8eb8df147f657830c8df34589875848663d
3
- size 243
 
 
 
 
distil-whisper_distil-large-v3_turbo_600MB/TextDecoder.mlmodelc/coremldata.bin DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:db10bd19b00bccc9b7d2a5a8317b6d73be16c01f78b3b3da47040ae604347b11
3
- size 633
 
 
 
 
distil-whisper_distil-large-v3_turbo_600MB/TextDecoder.mlmodelc/metadata.json DELETED
@@ -1,165 +0,0 @@
1
- [
2
- {
3
- "metadataOutputVersion" : "3.0",
4
- "storagePrecision" : "Float16",
5
- "outputSchema" : [
6
- {
7
- "hasShapeFlexibility" : "0",
8
- "isOptional" : "0",
9
- "dataType" : "Float16",
10
- "formattedType" : "MultiArray (Float16 1 × 1 × 51866)",
11
- "shortDescription" : "",
12
- "shape" : "[1, 1, 51866]",
13
- "name" : "logits",
14
- "type" : "MultiArray"
15
- },
16
- {
17
- "hasShapeFlexibility" : "0",
18
- "isOptional" : "0",
19
- "dataType" : "Float16",
20
- "formattedType" : "MultiArray (Float16 1 × 2560 × 1 × 1)",
21
- "shortDescription" : "",
22
- "shape" : "[1, 2560, 1, 1]",
23
- "name" : "key_cache_updates",
24
- "type" : "MultiArray"
25
- },
26
- {
27
- "hasShapeFlexibility" : "0",
28
- "isOptional" : "0",
29
- "dataType" : "Float16",
30
- "formattedType" : "MultiArray (Float16 1 × 2560 × 1 × 1)",
31
- "shortDescription" : "",
32
- "shape" : "[1, 2560, 1, 1]",
33
- "name" : "value_cache_updates",
34
- "type" : "MultiArray"
35
- },
36
- {
37
- "hasShapeFlexibility" : "0",
38
- "isOptional" : "0",
39
- "dataType" : "Float16",
40
- "formattedType" : "MultiArray (Float16 1 × 1500)",
41
- "shortDescription" : "",
42
- "shape" : "[1, 1500]",
43
- "name" : "alignment_heads_weights",
44
- "type" : "MultiArray"
45
- }
46
- ],
47
- "modelParameters" : [
48
-
49
- ],
50
- "specificationVersion" : 7,
51
- "mlProgramOperationTypeHistogram" : {
52
- "Split" : 2,
53
- "Concat" : 3,
54
- "Ios16.rsqrt" : 7,
55
- "Ios16.mul" : 26,
56
- "Squeeze" : 1,
57
- "SliceByIndex" : 40,
58
- "Ios16.sub" : 8,
59
- "Transpose" : 1,
60
- "Ios16.conv" : 20,
61
- "Ios16.add" : 20,
62
- "Ios16.linear" : 1,
63
- "Ios16.matmul" : 8,
64
- "Ios16.gelu" : 2,
65
- "Ios16.reduceMean" : 15,
66
- "ExpandDims" : 6,
67
- "Ios16.batchNorm" : 7,
68
- "Ios16.gather" : 2,
69
- "Ios16.reshape" : 16,
70
- "Ios16.softmax" : 4
71
- },
72
- "computePrecision" : "Mixed (Float16, Int32)",
73
- "isUpdatable" : "0",
74
- "availability" : {
75
- "macOS" : "13.0",
76
- "tvOS" : "16.0",
77
- "visionOS" : "1.0",
78
- "watchOS" : "9.0",
79
- "iOS" : "16.0",
80
- "macCatalyst" : "16.0"
81
- },
82
- "modelType" : {
83
- "name" : "MLModelType_mlProgram"
84
- },
85
- "userDefinedMetadata" : {
86
- "com.github.apple.coremltools.source_dialect" : "TorchScript",
87
- "com.github.apple.coremltools.source" : "torch==2.2.1",
88
- "com.github.apple.coremltools.version" : "7.1"
89
- },
90
- "inputSchema" : [
91
- {
92
- "hasShapeFlexibility" : "0",
93
- "isOptional" : "0",
94
- "dataType" : "Int32",
95
- "formattedType" : "MultiArray (Int32 1)",
96
- "shortDescription" : "",
97
- "shape" : "[1]",
98
- "name" : "input_ids",
99
- "type" : "MultiArray"
100
- },
101
- {
102
- "hasShapeFlexibility" : "0",
103
- "isOptional" : "0",
104
- "dataType" : "Int32",
105
- "formattedType" : "MultiArray (Int32 1)",
106
- "shortDescription" : "",
107
- "shape" : "[1]",
108
- "name" : "cache_length",
109
- "type" : "MultiArray"
110
- },
111
- {
112
- "hasShapeFlexibility" : "0",
113
- "isOptional" : "0",
114
- "dataType" : "Float16",
115
- "formattedType" : "MultiArray (Float16 1 × 2560 × 1 × 224)",
116
- "shortDescription" : "",
117
- "shape" : "[1, 2560, 1, 224]",
118
- "name" : "key_cache",
119
- "type" : "MultiArray"
120
- },
121
- {
122
- "hasShapeFlexibility" : "0",
123
- "isOptional" : "0",
124
- "dataType" : "Float16",
125
- "formattedType" : "MultiArray (Float16 1 × 2560 × 1 × 224)",
126
- "shortDescription" : "",
127
- "shape" : "[1, 2560, 1, 224]",
128
- "name" : "value_cache",
129
- "type" : "MultiArray"
130
- },
131
- {
132
- "hasShapeFlexibility" : "0",
133
- "isOptional" : "0",
134
- "dataType" : "Float16",
135
- "formattedType" : "MultiArray (Float16 1 × 224)",
136
- "shortDescription" : "",
137
- "shape" : "[1, 224]",
138
- "name" : "kv_cache_update_mask",
139
- "type" : "MultiArray"
140
- },
141
- {
142
- "hasShapeFlexibility" : "0",
143
- "isOptional" : "0",
144
- "dataType" : "Float16",
145
- "formattedType" : "MultiArray (Float16 1 × 1280 × 1 × 1500)",
146
- "shortDescription" : "",
147
- "shape" : "[1, 1280, 1, 1500]",
148
- "name" : "encoder_output_embeds",
149
- "type" : "MultiArray"
150
- },
151
- {
152
- "hasShapeFlexibility" : "0",
153
- "isOptional" : "0",
154
- "dataType" : "Float16",
155
- "formattedType" : "MultiArray (Float16 1 × 224)",
156
- "shortDescription" : "",
157
- "shape" : "[1, 224]",
158
- "name" : "decoder_key_padding_mask",
159
- "type" : "MultiArray"
160
- }
161
- ],
162
- "generatedClassName" : "TextDecoder",
163
- "method" : "predict"
164
- }
165
- ]
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
distil-whisper_distil-large-v3_turbo_600MB/TextDecoder.mlmodelc/model.mil DELETED
The diff for this file is too large to render. See raw diff
 
distil-whisper_distil-large-v3_turbo_600MB/TextDecoder.mlmodelc/weights/weight.bin DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:4e8e3906c4f1717fe848f6d45e6e052fc3d6048ec24d3a34e23e1c972b10bc84
3
- size 238986036
 
 
 
 
distil-whisper_distil-large-v3_turbo_600MB/TextDecoderContextPrefill.mlmodelc/analytics/coremldata.bin DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:177b3e51ce243362f79be85ee3d422665b49225bbb1993d72678a79a4788372e
3
- size 243
 
 
 
 
distil-whisper_distil-large-v3_turbo_600MB/TextDecoderContextPrefill.mlmodelc/coremldata.bin DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:e1fb28c51275e24a93535cf363012bbff21f6d486a8238d5fcf93d6288e2f781
3
- size 380
 
 
 
 
distil-whisper_distil-large-v3_turbo_600MB/TextDecoderContextPrefill.mlmodelc/metadata.json DELETED
@@ -1,82 +0,0 @@
1
- [
2
- {
3
- "metadataOutputVersion" : "3.0",
4
- "storagePrecision" : "Float16",
5
- "outputSchema" : [
6
- {
7
- "hasShapeFlexibility" : "0",
8
- "isOptional" : "0",
9
- "dataType" : "Float16",
10
- "formattedType" : "MultiArray (Float16 1 × 2560 × 1 × 3)",
11
- "shortDescription" : "",
12
- "shape" : "[1, 2560, 1, 3]",
13
- "name" : "key_cache_prefill",
14
- "type" : "MultiArray"
15
- },
16
- {
17
- "hasShapeFlexibility" : "0",
18
- "isOptional" : "0",
19
- "dataType" : "Float16",
20
- "formattedType" : "MultiArray (Float16 1 × 2560 × 1 × 3)",
21
- "shortDescription" : "",
22
- "shape" : "[1, 2560, 1, 3]",
23
- "name" : "value_cache_prefill",
24
- "type" : "MultiArray"
25
- }
26
- ],
27
- "modelParameters" : [
28
-
29
- ],
30
- "specificationVersion" : 8,
31
- "mlProgramOperationTypeHistogram" : {
32
- "Ios17.mul" : 1,
33
- "Ios17.cast" : 1,
34
- "Ios17.sub" : 1,
35
- "Ios17.reshape" : 2,
36
- "Ios17.add" : 1,
37
- "Ios17.gather" : 2
38
- },
39
- "computePrecision" : "Mixed (Float16, Int16, Int32)",
40
- "isUpdatable" : "0",
41
- "availability" : {
42
- "macOS" : "14.0",
43
- "tvOS" : "17.0",
44
- "visionOS" : "1.0",
45
- "watchOS" : "10.0",
46
- "iOS" : "17.0",
47
- "macCatalyst" : "17.0"
48
- },
49
- "modelType" : {
50
- "name" : "MLModelType_mlProgram"
51
- },
52
- "userDefinedMetadata" : {
53
- "com.github.apple.coremltools.source_dialect" : "TorchScript",
54
- "com.github.apple.coremltools.version" : "7.1",
55
- "com.github.apple.coremltools.source" : "torch==2.2.1"
56
- },
57
- "inputSchema" : [
58
- {
59
- "hasShapeFlexibility" : "0",
60
- "isOptional" : "0",
61
- "dataType" : "Int32",
62
- "formattedType" : "MultiArray (Int32 1)",
63
- "shortDescription" : "",
64
- "shape" : "[1]",
65
- "name" : "task",
66
- "type" : "MultiArray"
67
- },
68
- {
69
- "hasShapeFlexibility" : "0",
70
- "isOptional" : "0",
71
- "dataType" : "Int32",
72
- "formattedType" : "MultiArray (Int32 1)",
73
- "shortDescription" : "",
74
- "shape" : "[1]",
75
- "name" : "language",
76
- "type" : "MultiArray"
77
- }
78
- ],
79
- "generatedClassName" : "TextDecoderContextPrefill",
80
- "method" : "predict"
81
- }
82
- ]
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
distil-whisper_distil-large-v3_turbo_600MB/TextDecoderContextPrefill.mlmodelc/model.mil DELETED
@@ -1,27 +0,0 @@
1
- program(1.0)
2
- [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"}})]
3
- {
4
- func main<ios17>(tensor<int32, [1]> language, tensor<int32, [1]> task) {
5
- tensor<int32, []> var_6 = const()[name = tensor<string, []>("op_6"), val = tensor<int32, []>(50259)];
6
- tensor<int32, [1]> var_7 = sub(x = language, y = var_6)[name = tensor<string, []>("op_7")];
7
- tensor<int32, []> var_8 = const()[name = tensor<string, []>("op_8"), val = tensor<int32, []>(2)];
8
- tensor<int32, [1]> var_9 = mul(x = var_7, y = var_8)[name = tensor<string, []>("op_9")];
9
- tensor<int32, [1]> input = add(x = var_9, y = task)[name = tensor<string, []>("input")];
10
- tensor<int32, []> var_15_axis_0 = const()[name = tensor<string, []>("op_15_axis_0"), val = tensor<int32, []>(0)];
11
- tensor<int32, []> var_15_batch_dims_0 = const()[name = tensor<string, []>("op_15_batch_dims_0"), val = tensor<int32, []>(0)];
12
- tensor<bool, []> var_15_validate_indices_0 = const()[name = tensor<string, []>("op_15_validate_indices_0"), val = tensor<bool, []>(false)];
13
- tensor<fp16, [200, 7680]> key_cache_lut_weight_to_fp16 = const()[name = tensor<string, []>("key_cache_lut_weight_to_fp16"), val = tensor<fp16, [200, 7680]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(64)))];
14
- tensor<string, []> input_to_int16_dtype_0 = const()[name = tensor<string, []>("input_to_int16_dtype_0"), val = tensor<string, []>("int16")];
15
- tensor<int16, [1]> cast_6 = cast(dtype = input_to_int16_dtype_0, x = input)[name = tensor<string, []>("cast_6")];
16
- tensor<fp16, [1, 7680]> var_15_cast_fp16_cast_int16 = gather(axis = var_15_axis_0, batch_dims = var_15_batch_dims_0, indices = cast_6, validate_indices = var_15_validate_indices_0, x = key_cache_lut_weight_to_fp16)[name = tensor<string, []>("op_15_cast_fp16_cast_int16")];
17
- tensor<int32, [4]> var_20 = const()[name = tensor<string, []>("op_20"), val = tensor<int32, [4]>([1, 2560, 1, 3])];
18
- tensor<fp16, [1, 2560, 1, 3]> key_cache_prefill = reshape(shape = var_20, x = var_15_cast_fp16_cast_int16)[name = tensor<string, []>("op_21_cast_fp16")];
19
- tensor<int32, []> var_25_axis_0 = const()[name = tensor<string, []>("op_25_axis_0"), val = tensor<int32, []>(0)];
20
- tensor<int32, []> var_25_batch_dims_0 = const()[name = tensor<string, []>("op_25_batch_dims_0"), val = tensor<int32, []>(0)];
21
- tensor<bool, []> var_25_validate_indices_0 = const()[name = tensor<string, []>("op_25_validate_indices_0"), val = tensor<bool, []>(false)];
22
- tensor<fp16, [200, 7680]> value_cache_lut_weight_to_fp16 = const()[name = tensor<string, []>("value_cache_lut_weight_to_fp16"), val = tensor<fp16, [200, 7680]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(3072128)))];
23
- tensor<fp16, [1, 7680]> var_25_cast_fp16_cast_int16 = gather(axis = var_25_axis_0, batch_dims = var_25_batch_dims_0, indices = cast_6, validate_indices = var_25_validate_indices_0, x = value_cache_lut_weight_to_fp16)[name = tensor<string, []>("op_25_cast_fp16_cast_int16")];
24
- tensor<int32, [4]> var_30 = const()[name = tensor<string, []>("op_30"), val = tensor<int32, [4]>([1, 2560, 1, 3])];
25
- tensor<fp16, [1, 2560, 1, 3]> value_cache_prefill = reshape(shape = var_30, x = var_25_cast_fp16_cast_int16)[name = tensor<string, []>("op_31_cast_fp16")];
26
- } -> (key_cache_prefill, value_cache_prefill);
27
- }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
distil-whisper_distil-large-v3_turbo_600MB/TextDecoderContextPrefill.mlmodelc/weights/weight.bin DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:26266f85b3830f13b43e4fec705e3999329b1af171b23f6eff4e89fe9c137ae0
3
- size 6144192
 
 
 
 
distil-whisper_distil-large-v3_turbo_600MB/config.json DELETED
@@ -1 +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_turbo_600MB/generation_config.json DELETED
@@ -1 +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"}