arda-argmax commited on
Commit
3ddc513
1 Parent(s): 5c268ba

Upload model openai_whisper-large-v2_949MB - QLoRA compressed encoder decoder variant

Browse files
openai_whisper-large-v2_949MB/AudioEncoder.mlmodelc/analytics/coremldata.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:18a25d3e04409cb8d50b431eb55a29a02c10e2309756e506953472a1e74599fd
3
+ size 243
openai_whisper-large-v2_949MB/AudioEncoder.mlmodelc/coremldata.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7db300c721e92649f92a98e717a322622a533c624889493f104a2e1014736310
3
+ size 347
openai_whisper-large-v2_949MB/AudioEncoder.mlmodelc/metadata.json ADDED
@@ -0,0 +1,68 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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" : 162,
23
+ "Ios17.rsqrt" : 65,
24
+ "Ios16.constexprLutToDense" : 192,
25
+ "Ios17.conv" : 578,
26
+ "Ios17.matmul" : 64,
27
+ "Ios17.sub" : 65,
28
+ "Ios17.add" : 322,
29
+ "Ios16.reduceMean" : 130,
30
+ "Ios17.batchNorm" : 65,
31
+ "Ios16.softmax" : 32,
32
+ "Ios16.gelu" : 34,
33
+ "Ios17.reshape" : 128
34
+ },
35
+ "computePrecision" : "Mixed (Float16, Float32, Int32)",
36
+ "isUpdatable" : "0",
37
+ "availability" : {
38
+ "macOS" : "14.0",
39
+ "tvOS" : "17.0",
40
+ "visionOS" : "1.0",
41
+ "watchOS" : "10.0",
42
+ "iOS" : "17.0",
43
+ "macCatalyst" : "17.0"
44
+ },
45
+ "modelType" : {
46
+ "name" : "MLModelType_mlProgram"
47
+ },
48
+ "userDefinedMetadata" : {
49
+ "com.github.apple.coremltools.source_dialect" : "TorchScript",
50
+ "com.github.apple.coremltools.source" : "torch==2.2.1",
51
+ "com.github.apple.coremltools.version" : "7.1"
52
+ },
53
+ "inputSchema" : [
54
+ {
55
+ "hasShapeFlexibility" : "0",
56
+ "isOptional" : "0",
57
+ "dataType" : "Float16",
58
+ "formattedType" : "MultiArray (Float16 1 × 80 × 1 × 3000)",
59
+ "shortDescription" : "",
60
+ "shape" : "[1, 80, 1, 3000]",
61
+ "name" : "melspectrogram_features",
62
+ "type" : "MultiArray"
63
+ }
64
+ ],
65
+ "generatedClassName" : "AudioEncoder_mixedBitPalettized_4_0_bit",
66
+ "method" : "predict"
67
+ }
68
+ ]
openai_whisper-large-v2_949MB/AudioEncoder.mlmodelc/model.mil ADDED
The diff for this file is too large to render. See raw diff
 
openai_whisper-large-v2_949MB/AudioEncoder.mlmodelc/weights/weight.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9cde1ffb6ef9d99026aa70bcd089dd58644e5a25a5b701f106d6e788872d7d8b
3
+ size 353539776
openai_whisper-large-v2_949MB/MelSpectrogram.mlmodelc/analytics/coremldata.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:562a52640a05fc46d336dde5d9a2a32f518e8a501e8584167699ed96ffb0e9df
3
+ size 243
openai_whisper-large-v2_949MB/MelSpectrogram.mlmodelc/coremldata.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4f4a237d5012a4714e8ed606014d849b789ac7a557c5a2142e6fba0ef4100fdc
3
+ size 328
openai_whisper-large-v2_949MB/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" : 8,
21
+ "mlProgramOperationTypeHistogram" : {
22
+ "Ios17.mul" : 2,
23
+ "Ios17.sub" : 1,
24
+ "Ios17.matmul" : 1,
25
+ "Ios17.conv" : 2,
26
+ "Ios17.log" : 1,
27
+ "Ios17.add" : 3,
28
+ "Ios16.square" : 2,
29
+ "Ios17.sliceByIndex" : 1,
30
+ "Ios17.expandDims" : 4,
31
+ "Ios17.maximum" : 1,
32
+ "Ios16.reduceMax" : 1,
33
+ "Ios17.squeeze" : 2,
34
+ "Ios17.reshape" : 2,
35
+ "Identity" : 1,
36
+ "Pad" : 1
37
+ },
38
+ "computePrecision" : "Mixed (Float16, Float32, Int32)",
39
+ "isUpdatable" : "0",
40
+ "availability" : {
41
+ "macOS" : "14.0",
42
+ "tvOS" : "17.0",
43
+ "visionOS" : "1.0",
44
+ "watchOS" : "10.0",
45
+ "iOS" : "17.0",
46
+ "macCatalyst" : "17.0"
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-large-v2_949MB/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<ios17>(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<fp32, []> log_0_epsilon_0 = const()[name = tensor<string, []>("log_0_epsilon_0"), val = tensor<fp32, []>(0x1p-149)];
49
+ tensor<fp16, [80, 3000]> log_0_cast_fp16 = log(epsilon = log_0_epsilon_0, 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-large-v2_949MB/MelSpectrogram.mlmodelc/weights/weight.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:57ccef11a468a4d409f3d6c236b110d99bb4b1e295614a6e6d7387bf0cecbbcf
3
+ size 354080
openai_whisper-large-v2_949MB/TextDecoder.mlmodelc/analytics/coremldata.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:730b41c2ace4933aedb00ec573e183fc404c5da34f9d2650a8401e1f67066899
3
+ size 243
openai_whisper-large-v2_949MB/TextDecoder.mlmodelc/coremldata.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:211920c257a33cf3fb261fdb310a816dc88e0d2403ce6d3e2308a23b1647f720
3
+ size 637
openai_whisper-large-v2_949MB/TextDecoder.mlmodelc/metadata.json ADDED
@@ -0,0 +1,167 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [
2
+ {
3
+ "metadataOutputVersion" : "3.0",
4
+ "storagePrecision" : "Mixed (Float16, Palettized (4 bits), Palettized (6 bits))",
5
+ "outputSchema" : [
6
+ {
7
+ "hasShapeFlexibility" : "0",
8
+ "isOptional" : "0",
9
+ "dataType" : "Float16",
10
+ "formattedType" : "MultiArray (Float16 1 × 1 × 51865)",
11
+ "shortDescription" : "",
12
+ "shape" : "[1, 1, 51865]",
13
+ "name" : "logits",
14
+ "type" : "MultiArray"
15
+ },
16
+ {
17
+ "hasShapeFlexibility" : "0",
18
+ "isOptional" : "0",
19
+ "dataType" : "Float16",
20
+ "formattedType" : "MultiArray (Float16 1 × 40960 × 1 × 1)",
21
+ "shortDescription" : "",
22
+ "shape" : "[1, 40960, 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 × 40960 × 1 × 1)",
31
+ "shortDescription" : "",
32
+ "shape" : "[1, 40960, 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" : 8,
51
+ "mlProgramOperationTypeHistogram" : {
52
+ "Ios17.reshape" : 256,
53
+ "Ios16.reduceMean" : 195,
54
+ "Ios16.softmax" : 64,
55
+ "Ios17.matmul" : 128,
56
+ "Ios17.transpose" : 1,
57
+ "Split" : 2,
58
+ "Ios17.expandDims" : 6,
59
+ "Ios17.add" : 610,
60
+ "Ios17.sliceByIndex" : 46,
61
+ "Ios17.squeeze" : 1,
62
+ "Ios17.gather" : 2,
63
+ "Ios17.batchNorm" : 97,
64
+ "Ios16.gelu" : 32,
65
+ "Ios17.cast" : 1,
66
+ "Ios16.constexprLutToDense" : 320,
67
+ "Ios17.conv" : 960,
68
+ "Ios17.sub" : 98,
69
+ "Ios17.concat" : 3,
70
+ "Ios17.linear" : 1,
71
+ "Ios17.rsqrt" : 97,
72
+ "Ios17.mul" : 386
73
+ },
74
+ "computePrecision" : "Mixed (Float16, Float32, Int16, Int32)",
75
+ "isUpdatable" : "0",
76
+ "availability" : {
77
+ "macOS" : "14.0",
78
+ "tvOS" : "17.0",
79
+ "visionOS" : "1.0",
80
+ "watchOS" : "10.0",
81
+ "iOS" : "17.0",
82
+ "macCatalyst" : "17.0"
83
+ },
84
+ "modelType" : {
85
+ "name" : "MLModelType_mlProgram"
86
+ },
87
+ "userDefinedMetadata" : {
88
+ "com.github.apple.coremltools.source_dialect" : "TorchScript",
89
+ "com.github.apple.coremltools.source" : "torch==2.2.1",
90
+ "com.github.apple.coremltools.version" : "7.1"
91
+ },
92
+ "inputSchema" : [
93
+ {
94
+ "hasShapeFlexibility" : "0",
95
+ "isOptional" : "0",
96
+ "dataType" : "Int32",
97
+ "formattedType" : "MultiArray (Int32 1)",
98
+ "shortDescription" : "",
99
+ "shape" : "[1]",
100
+ "name" : "input_ids",
101
+ "type" : "MultiArray"
102
+ },
103
+ {
104
+ "hasShapeFlexibility" : "0",
105
+ "isOptional" : "0",
106
+ "dataType" : "Int32",
107
+ "formattedType" : "MultiArray (Int32 1)",
108
+ "shortDescription" : "",
109
+ "shape" : "[1]",
110
+ "name" : "cache_length",
111
+ "type" : "MultiArray"
112
+ },
113
+ {
114
+ "hasShapeFlexibility" : "0",
115
+ "isOptional" : "0",
116
+ "dataType" : "Float16",
117
+ "formattedType" : "MultiArray (Float16 1 × 40960 × 1 × 448)",
118
+ "shortDescription" : "",
119
+ "shape" : "[1, 40960, 1, 448]",
120
+ "name" : "key_cache",
121
+ "type" : "MultiArray"
122
+ },
123
+ {
124
+ "hasShapeFlexibility" : "0",
125
+ "isOptional" : "0",
126
+ "dataType" : "Float16",
127
+ "formattedType" : "MultiArray (Float16 1 × 40960 × 1 × 448)",
128
+ "shortDescription" : "",
129
+ "shape" : "[1, 40960, 1, 448]",
130
+ "name" : "value_cache",
131
+ "type" : "MultiArray"
132
+ },
133
+ {
134
+ "hasShapeFlexibility" : "0",
135
+ "isOptional" : "0",
136
+ "dataType" : "Float16",
137
+ "formattedType" : "MultiArray (Float16 1 × 448)",
138
+ "shortDescription" : "",
139
+ "shape" : "[1, 448]",
140
+ "name" : "kv_cache_update_mask",
141
+ "type" : "MultiArray"
142
+ },
143
+ {
144
+ "hasShapeFlexibility" : "0",
145
+ "isOptional" : "0",
146
+ "dataType" : "Float16",
147
+ "formattedType" : "MultiArray (Float16 1 × 1280 × 1 × 1500)",
148
+ "shortDescription" : "",
149
+ "shape" : "[1, 1280, 1, 1500]",
150
+ "name" : "encoder_output_embeds",
151
+ "type" : "MultiArray"
152
+ },
153
+ {
154
+ "hasShapeFlexibility" : "0",
155
+ "isOptional" : "0",
156
+ "dataType" : "Float16",
157
+ "formattedType" : "MultiArray (Float16 1 × 448)",
158
+ "shortDescription" : "",
159
+ "shape" : "[1, 448]",
160
+ "name" : "decoder_key_padding_mask",
161
+ "type" : "MultiArray"
162
+ }
163
+ ],
164
+ "generatedClassName" : "TextDecoder_mixedBitPalettized_4_0_bit",
165
+ "method" : "predict"
166
+ }
167
+ ]
openai_whisper-large-v2_949MB/TextDecoder.mlmodelc/model.mil ADDED
The diff for this file is too large to render. See raw diff
 
openai_whisper-large-v2_949MB/TextDecoder.mlmodelc/weights/weight.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:87dbbb2505b11feebce23f3bb044eedbd4d90bbdef06cda40d63481a0362b6bf
3
+ size 592355826
openai_whisper-large-v2_949MB/config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"_name_or_path": "openai/whisper-large-v2", "activation_dropout": 0.0, "activation_function": "gelu", "architectures": ["WhisperForConditionalGeneration"], "attention_dropout": 0.0, "begin_suppress_tokens": [220, 50257], "bos_token_id": 50257, "d_model": 1280, "decoder_attention_heads": 20, "decoder_ffn_dim": 5120, "decoder_layerdrop": 0.0, "decoder_layers": 32, "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, "forced_decoder_ids": [[1, 50259], [2, 50359], [3, 50363]], "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": 32, "num_mel_bins": 80, "pad_token_id": 50257, "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, 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, 50358, 50359, 50360, 50361, 50362], "torch_dtype": "float32", "transformers_version": "4.27.0.dev0", "use_cache": true, "vocab_size": 51865}
openai_whisper-large-v2_949MB/generation_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"alignment_heads": [[10, 12], [13, 17], [16, 11], [16, 12], [16, 13], [17, 15], [17, 16], [18, 4], [18, 11], [18, 19], [19, 11], [21, 2], [21, 3], [22, 3], [22, 9], [22, 12], [23, 5], [23, 7], [23, 13], [25, 5], [26, 1], [26, 12], [27, 15]], "begin_suppress_tokens": [220, 50257], "bos_token_id": 50257, "decoder_start_token_id": 50258, "eos_token_id": 50257, "forced_decoder_ids": [[1, null], [2, 50359]], "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, "<|zh|>": 50260}, "max_initial_timestamp_index": 50, "max_length": 448, "no_timestamps_token_id": 50363, "pad_token_id": 50257, "prev_sot_token_id": 50361, "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, 50358, 50359, 50360, 50361, 50362], "task_to_id": {"transcribe": 50359, "translate": 50358}, "transformers_version": "4.31.0.dev0"}