whisperkittools-192190cedeefc4d317d13c2196dc29f9a9f99628 generated files: openai_whisper-base.en
Browse files- openai_whisper-base.en/AudioEncoder.mlmodelc/analytics/coremldata.bin +3 -0
- openai_whisper-base.en/AudioEncoder.mlmodelc/coremldata.bin +3 -0
- openai_whisper-base.en/AudioEncoder.mlmodelc/metadata.json +69 -0
- openai_whisper-base.en/AudioEncoder.mlmodelc/model.mil +0 -0
- openai_whisper-base.en/AudioEncoder.mlmodelc/weights/weight.bin +3 -0
- openai_whisper-base.en/MelSpectrogram.mlmodelc/analytics/coremldata.bin +3 -0
- openai_whisper-base.en/MelSpectrogram.mlmodelc/coremldata.bin +3 -0
- openai_whisper-base.en/MelSpectrogram.mlmodelc/metadata.json +71 -0
- openai_whisper-base.en/MelSpectrogram.mlmodelc/model.mil +66 -0
- openai_whisper-base.en/MelSpectrogram.mlmodelc/weights/weight.bin +3 -0
- openai_whisper-base.en/TextDecoder.mlmodelc/analytics/coremldata.bin +3 -0
- openai_whisper-base.en/TextDecoder.mlmodelc/coremldata.bin +3 -0
- openai_whisper-base.en/TextDecoder.mlmodelc/metadata.json +165 -0
- openai_whisper-base.en/TextDecoder.mlmodelc/model.mil +0 -0
- openai_whisper-base.en/TextDecoder.mlmodelc/weights/weight.bin +3 -0
- openai_whisper-base.en/config.json +1 -0
- openai_whisper-base.en/generation_config.json +1 -0
openai_whisper-base.en/AudioEncoder.mlmodelc/analytics/coremldata.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c4e096b2abd561f00b9b698401df3fbe1a0d0c8d2476ff19cb4e1995680e827e
|
3 |
+
size 243
|
openai_whisper-base.en/AudioEncoder.mlmodelc/coremldata.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e316980638e2099e83cb1a93b903717dc12b3c2168d0ab69113764c3767696ba
|
3 |
+
size 347
|
openai_whisper-base.en/AudioEncoder.mlmodelc/metadata.json
ADDED
@@ -0,0 +1,69 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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 × 512 × 1 × 1500)",
|
11 |
+
"shortDescription" : "",
|
12 |
+
"shape" : "[1, 512, 1, 1500]",
|
13 |
+
"name" : "encoder_output_embeds",
|
14 |
+
"type" : "MultiArray"
|
15 |
+
}
|
16 |
+
],
|
17 |
+
"modelParameters" : [
|
18 |
+
|
19 |
+
],
|
20 |
+
"specificationVersion" : 7,
|
21 |
+
"mlProgramOperationTypeHistogram" : {
|
22 |
+
"Concat" : 54,
|
23 |
+
"Ios16.rsqrt" : 13,
|
24 |
+
"Ios16.mul" : 218,
|
25 |
+
"SliceByIndex" : 336,
|
26 |
+
"Ios16.sub" : 13,
|
27 |
+
"Transpose" : 6,
|
28 |
+
"Ios16.einsum" : 384,
|
29 |
+
"Ios16.conv" : 38,
|
30 |
+
"Ios16.add" : 26,
|
31 |
+
"Ios16.reduceMean" : 26,
|
32 |
+
"Ios16.softmax" : 192,
|
33 |
+
"Ios16.gelu" : 8,
|
34 |
+
"Ios16.batchNorm" : 13
|
35 |
+
},
|
36 |
+
"computePrecision" : "Mixed (Float16, Int32)",
|
37 |
+
"isUpdatable" : "0",
|
38 |
+
"availability" : {
|
39 |
+
"macOS" : "13.0",
|
40 |
+
"tvOS" : "16.0",
|
41 |
+
"visionOS" : "1.0",
|
42 |
+
"watchOS" : "9.0",
|
43 |
+
"iOS" : "16.0",
|
44 |
+
"macCatalyst" : "16.0"
|
45 |
+
},
|
46 |
+
"modelType" : {
|
47 |
+
"name" : "MLModelType_mlProgram"
|
48 |
+
},
|
49 |
+
"userDefinedMetadata" : {
|
50 |
+
"com.github.apple.coremltools.source_dialect" : "TorchScript",
|
51 |
+
"com.github.apple.coremltools.source" : "torch==2.2.1",
|
52 |
+
"com.github.apple.coremltools.version" : "7.1"
|
53 |
+
},
|
54 |
+
"inputSchema" : [
|
55 |
+
{
|
56 |
+
"hasShapeFlexibility" : "0",
|
57 |
+
"isOptional" : "0",
|
58 |
+
"dataType" : "Float16",
|
59 |
+
"formattedType" : "MultiArray (Float16 1 × 80 × 1 × 3000)",
|
60 |
+
"shortDescription" : "",
|
61 |
+
"shape" : "[1, 80, 1, 3000]",
|
62 |
+
"name" : "melspectrogram_features",
|
63 |
+
"type" : "MultiArray"
|
64 |
+
}
|
65 |
+
],
|
66 |
+
"generatedClassName" : "AudioEncoder",
|
67 |
+
"method" : "predict"
|
68 |
+
}
|
69 |
+
]
|
openai_whisper-base.en/AudioEncoder.mlmodelc/model.mil
ADDED
The diff for this file is too large to render.
See raw diff
|
|
openai_whisper-base.en/AudioEncoder.mlmodelc/weights/weight.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d0490f39cdf5efd2bd955802797391658ca8dbfd24a641f9cb1932a274444f49
|
3 |
+
size 41189632
|
openai_whisper-base.en/MelSpectrogram.mlmodelc/analytics/coremldata.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7f77e6457285248f99cd7aa3fd4cc2efbb17733e63e7023ac53abe1f95785d07
|
3 |
+
size 243
|
openai_whisper-base.en/MelSpectrogram.mlmodelc/coremldata.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:dabdc5aa69f6ef4d97dc9499f5c30514e00e96b53b750b33a5a6471363c71662
|
3 |
+
size 328
|
openai_whisper-base.en/MelSpectrogram.mlmodelc/metadata.json
ADDED
@@ -0,0 +1,71 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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 × 80 × 1 × 3000)",
|
11 |
+
"shortDescription" : "",
|
12 |
+
"shape" : "[1, 80, 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 |
+
]
|
openai_whisper-base.en/MelSpectrogram.mlmodelc/model.mil
ADDED
@@ -0,0 +1,66 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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, [80, 201]> mel_filters_to_fp16 = const()[name = tensor<string, []>("mel_filters_to_fp16"), val = tensor<fp16, [80, 201]>(BLOBFILE(path = tensor<string, []>("@model_path/weights/weight.bin"), offset = tensor<uint64, []>(321856)))];
|
45 |
+
tensor<fp16, [80, 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, [80, 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, [80, 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, [80, 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, [80, 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, [80, 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, [80, 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, 80, 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, 80, 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 |
+
}
|
openai_whisper-base.en/MelSpectrogram.mlmodelc/weights/weight.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d34d5577f79052c621cfcd7ed2e83e03b73e81c7a5f78a8dc9fa20884970483e
|
3 |
+
size 354080
|
openai_whisper-base.en/TextDecoder.mlmodelc/analytics/coremldata.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e870aac5de88676c1ad93cb3939f257e6de30e5235c3f9d7686398ebecafdb8d
|
3 |
+
size 243
|
openai_whisper-base.en/TextDecoder.mlmodelc/coremldata.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4b23ff9d9fe376f2b86f9cf55af3218560a4e4d692041360510fbad490c0db43
|
3 |
+
size 633
|
openai_whisper-base.en/TextDecoder.mlmodelc/metadata.json
ADDED
@@ -0,0 +1,165 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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 × 51864)",
|
11 |
+
"shortDescription" : "",
|
12 |
+
"shape" : "[1, 1, 51864]",
|
13 |
+
"name" : "logits",
|
14 |
+
"type" : "MultiArray"
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"hasShapeFlexibility" : "0",
|
18 |
+
"isOptional" : "0",
|
19 |
+
"dataType" : "Float16",
|
20 |
+
"formattedType" : "MultiArray (Float16 1 × 3072 × 1 × 1)",
|
21 |
+
"shortDescription" : "",
|
22 |
+
"shape" : "[1, 3072, 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 × 3072 × 1 × 1)",
|
31 |
+
"shortDescription" : "",
|
32 |
+
"shape" : "[1, 3072, 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" : 19,
|
55 |
+
"Ios16.mul" : 74,
|
56 |
+
"Squeeze" : 1,
|
57 |
+
"SliceByIndex" : 10,
|
58 |
+
"Ios16.sub" : 20,
|
59 |
+
"Transpose" : 1,
|
60 |
+
"Ios16.conv" : 60,
|
61 |
+
"Ios16.add" : 56,
|
62 |
+
"Ios16.linear" : 1,
|
63 |
+
"Ios16.matmul" : 24,
|
64 |
+
"Ios16.gelu" : 6,
|
65 |
+
"Ios16.reduceMean" : 39,
|
66 |
+
"ExpandDims" : 6,
|
67 |
+
"Ios16.batchNorm" : 19,
|
68 |
+
"Ios16.gather" : 2,
|
69 |
+
"Ios16.reshape" : 48,
|
70 |
+
"Ios16.softmax" : 12
|
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.version" : "7.1",
|
88 |
+
"com.github.apple.coremltools.source" : "torch==2.2.1"
|
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 × 3072 × 1 × 224)",
|
116 |
+
"shortDescription" : "",
|
117 |
+
"shape" : "[1, 3072, 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 × 3072 × 1 × 224)",
|
126 |
+
"shortDescription" : "",
|
127 |
+
"shape" : "[1, 3072, 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 × 512 × 1 × 1500)",
|
146 |
+
"shortDescription" : "",
|
147 |
+
"shape" : "[1, 512, 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 |
+
]
|
openai_whisper-base.en/TextDecoder.mlmodelc/model.mil
ADDED
The diff for this file is too large to render.
See raw diff
|
|
openai_whisper-base.en/TextDecoder.mlmodelc/weights/weight.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f6becbf52a2c0e68a006e05e50a86fc9e4ab1619abfa114314014a05f37dbc18
|
3 |
+
size 104121136
|
openai_whisper-base.en/config.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"_name_or_path": "openai/whisper-base.en", "activation_dropout": 0.0, "activation_function": "gelu", "architectures": ["WhisperForConditionalGeneration"], "attention_dropout": 0.0, "begin_suppress_tokens": [220, 50256], "bos_token_id": 50257, "d_model": 512, "decoder_attention_heads": 8, "decoder_ffn_dim": 2048, "decoder_layerdrop": 0.0, "decoder_layers": 6, "decoder_start_token_id": 50257, "dropout": 0.0, "encoder_attention_heads": 8, "encoder_ffn_dim": 2048, "encoder_layerdrop": 0.0, "encoder_layers": 6, "eos_token_id": 50256, "forced_decoder_ids": [[1, 50362]], "init_std": 0.02, "is_encoder_decoder": true, "max_length": 448, "max_source_positions": 1500, "max_target_positions": 448, "model_type": "whisper", "num_hidden_layers": 6, "num_mel_bins": 80, "pad_token_id": 50256, "scale_embedding": 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, 357, 366, 438, 532, 685, 705, 796, 930, 1058, 1220, 1267, 1279, 1303, 1343, 1377, 1391, 1635, 1782, 1875, 2162, 2361, 2488, 3467, 4008, 4211, 4600, 4808, 5299, 5855, 6329, 7203, 9609, 9959, 10563, 10786, 11420, 11709, 11907, 13163, 13697, 13700, 14808, 15306, 16410, 16791, 17992, 19203, 19510, 20724, 22305, 22935, 27007, 30109, 30420, 33409, 34949, 40283, 40493, 40549, 47282, 49146, 50257, 50357, 50358, 50359, 50360, 50361], "torch_dtype": "float32", "transformers_version": "4.27.0.dev0", "use_cache": true, "vocab_size": 51864}
|
openai_whisper-base.en/generation_config.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"alignment_heads": [[3, 3], [4, 7], [5, 1], [5, 5], [5, 7]], "begin_suppress_tokens": [220, 50256], "bos_token_id": 50257, "decoder_start_token_id": 50257, "eos_token_id": 50256, "forced_decoder_ids": [[1, 50362]], "is_multilingual": false, "max_initial_timestamp_index": 50, "max_length": 448, "no_timestamps_token_id": 50362, "pad_token_id": 50256, "prev_sot_token_id": 50360, "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, 357, 366, 438, 532, 685, 705, 796, 930, 1058, 1220, 1267, 1279, 1303, 1343, 1377, 1391, 1635, 1782, 1875, 2162, 2361, 2488, 3467, 4008, 4211, 4600, 4808, 5299, 5855, 6329, 7203, 9609, 9959, 10563, 10786, 11420, 11709, 11907, 13163, 13697, 13700, 14808, 15306, 16410, 16791, 17992, 19203, 19510, 20724, 22305, 22935, 27007, 30109, 30420, 33409, 34949, 40283, 40493, 40549, 47282, 49146, 50257, 50357, 50358, 50359, 50360, 50361], "transformers_version": "4.31.0.dev0"}
|