mobi commited on
Commit
7d46ea5
1 Parent(s): 1f15d17

Upload folder using huggingface_hub

Browse files
.gitattributes CHANGED
@@ -25,7 +25,6 @@
25
  *.safetensors filter=lfs diff=lfs merge=lfs -text
26
  saved_model/**/* filter=lfs diff=lfs merge=lfs -text
27
  *.tar.* filter=lfs diff=lfs merge=lfs -text
28
- *.tar filter=lfs diff=lfs merge=lfs -text
29
  *.tflite filter=lfs diff=lfs merge=lfs -text
30
  *.tgz filter=lfs diff=lfs merge=lfs -text
31
  *.wasm filter=lfs diff=lfs merge=lfs -text
 
25
  *.safetensors filter=lfs diff=lfs merge=lfs -text
26
  saved_model/**/* filter=lfs diff=lfs merge=lfs -text
27
  *.tar.* filter=lfs diff=lfs merge=lfs -text
 
28
  *.tflite filter=lfs diff=lfs merge=lfs -text
29
  *.tgz filter=lfs diff=lfs merge=lfs -text
30
  *.wasm filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,49 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ datasets:
4
+ - Murple/ksponspeech
5
+ language:
6
+ - ko
7
+ metrics:
8
+ - cer
9
+ - wer
10
+ pipeline_tag: automatic-speech-recognition
11
+ ---
12
+ # Whisper-Medium-KsponSpeech
13
+
14
+ The Whisper-medium Model finetunned with [KsponSpeech](https://huggingface.co/datasets/Murple/ksponspeech)
15
+
16
+
17
+ ### Model Description
18
+
19
+ <!-- Provide a longer summary of what this model is. -->
20
+
21
+
22
+
23
+ - **Developed by :** [yw0nam](https://github.com/yw0nam)
24
+ - **Shared by :** [yw0nam](https://github.com/yw0nam)
25
+ - **Model type :** ASR
26
+ - **License:** [apache-2.0]
27
+
28
+ ## Uses
29
+
30
+ ```
31
+
32
+ processor = WhisperProcessor.from_pretrained("openai/whisper-medium", language="ko", task="transcribe")
33
+ model = WhisperForConditionalGeneration.from_pretrained('spow12/whisper-medium-zeroth_korean').cuda()
34
+
35
+ data, _ = librosa.load(wav_path, sr=16000)
36
+ input_features = processor(data, sampling_rate=16000, return_tensors="pt").input_features.cuda()
37
+
38
+ predicted_ids = model.generate(input_features)
39
+ transcription = processor.batch_decode(predicted_ids, skip_special_tokens=True)[0]
40
+
41
+ ```
42
+
43
+ ### Metrics
44
+
45
+ Metric | result |
46
+ --- | --- |
47
+ WER | 3.96 |
48
+ CER | 1.71 |
49
+
config.json ADDED
@@ -0,0 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "./models_zoo/openai_whisper-medium_zeroth_korean/model_weights/",
3
+ "activation_dropout": 0.0,
4
+ "activation_function": "gelu",
5
+ "apply_spec_augment": false,
6
+ "architectures": [
7
+ "WhisperForConditionalGeneration"
8
+ ],
9
+ "attention_dropout": 0.0,
10
+ "begin_suppress_tokens": [
11
+ 220,
12
+ 50257
13
+ ],
14
+ "bos_token_id": 50257,
15
+ "classifier_proj_size": 256,
16
+ "d_model": 1024,
17
+ "decoder_attention_heads": 16,
18
+ "decoder_ffn_dim": 4096,
19
+ "decoder_layerdrop": 0.0,
20
+ "decoder_layers": 24,
21
+ "decoder_start_token_id": 50258,
22
+ "dropout": 0.0,
23
+ "encoder_attention_heads": 16,
24
+ "encoder_ffn_dim": 4096,
25
+ "encoder_layerdrop": 0.0,
26
+ "encoder_layers": 24,
27
+ "eos_token_id": 50257,
28
+ "forced_decoder_ids": [
29
+ [
30
+ 1,
31
+ 50264
32
+ ],
33
+ [
34
+ 2,
35
+ 50359
36
+ ],
37
+ [
38
+ 3,
39
+ 50363
40
+ ]
41
+ ],
42
+ "init_std": 0.02,
43
+ "is_encoder_decoder": true,
44
+ "mask_feature_length": 10,
45
+ "mask_feature_min_masks": 0,
46
+ "mask_feature_prob": 0.0,
47
+ "mask_time_length": 10,
48
+ "mask_time_min_masks": 2,
49
+ "mask_time_prob": 0.05,
50
+ "max_length": 448,
51
+ "max_source_positions": 1500,
52
+ "max_target_positions": 448,
53
+ "model_type": "whisper",
54
+ "num_hidden_layers": 24,
55
+ "num_mel_bins": 80,
56
+ "pad_token_id": 50257,
57
+ "scale_embedding": false,
58
+ "suppress_tokens": [],
59
+ "torch_dtype": "float32",
60
+ "transformers_version": "4.27.4",
61
+ "use_cache": true,
62
+ "use_weighted_layer_sum": false,
63
+ "vocab_size": 51865
64
+ }
generation_config.json ADDED
@@ -0,0 +1,225 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "begin_suppress_tokens": [
3
+ 220,
4
+ 50257
5
+ ],
6
+ "bos_token_id": 50257,
7
+ "decoder_start_token_id": 50258,
8
+ "eos_token_id": 50257,
9
+ "forced_decoder_ids": [
10
+ [
11
+ 1,
12
+ 50264
13
+ ],
14
+ [
15
+ 2,
16
+ 50359
17
+ ],
18
+ [
19
+ 3,
20
+ 50363
21
+ ]
22
+ ],
23
+ "is_multilingual": true,
24
+ "lang_to_id": {
25
+ "<|af|>": 50327,
26
+ "<|am|>": 50334,
27
+ "<|ar|>": 50272,
28
+ "<|as|>": 50350,
29
+ "<|az|>": 50304,
30
+ "<|ba|>": 50355,
31
+ "<|be|>": 50330,
32
+ "<|bg|>": 50292,
33
+ "<|bn|>": 50302,
34
+ "<|bo|>": 50347,
35
+ "<|br|>": 50309,
36
+ "<|bs|>": 50315,
37
+ "<|ca|>": 50270,
38
+ "<|cs|>": 50283,
39
+ "<|cy|>": 50297,
40
+ "<|da|>": 50285,
41
+ "<|de|>": 50261,
42
+ "<|el|>": 50281,
43
+ "<|en|>": 50259,
44
+ "<|es|>": 50262,
45
+ "<|et|>": 50307,
46
+ "<|eu|>": 50310,
47
+ "<|fa|>": 50300,
48
+ "<|fi|>": 50277,
49
+ "<|fo|>": 50338,
50
+ "<|fr|>": 50265,
51
+ "<|gl|>": 50319,
52
+ "<|gu|>": 50333,
53
+ "<|haw|>": 50352,
54
+ "<|ha|>": 50354,
55
+ "<|he|>": 50279,
56
+ "<|hi|>": 50276,
57
+ "<|hr|>": 50291,
58
+ "<|ht|>": 50339,
59
+ "<|hu|>": 50286,
60
+ "<|hy|>": 50312,
61
+ "<|id|>": 50275,
62
+ "<|is|>": 50311,
63
+ "<|it|>": 50274,
64
+ "<|ja|>": 50266,
65
+ "<|jw|>": 50356,
66
+ "<|ka|>": 50329,
67
+ "<|kk|>": 50316,
68
+ "<|km|>": 50323,
69
+ "<|kn|>": 50306,
70
+ "<|ko|>": 50264,
71
+ "<|la|>": 50294,
72
+ "<|lb|>": 50345,
73
+ "<|ln|>": 50353,
74
+ "<|lo|>": 50336,
75
+ "<|lt|>": 50293,
76
+ "<|lv|>": 50301,
77
+ "<|mg|>": 50349,
78
+ "<|mi|>": 50295,
79
+ "<|mk|>": 50308,
80
+ "<|ml|>": 50296,
81
+ "<|mn|>": 50314,
82
+ "<|mr|>": 50320,
83
+ "<|ms|>": 50282,
84
+ "<|mt|>": 50343,
85
+ "<|my|>": 50346,
86
+ "<|ne|>": 50313,
87
+ "<|nl|>": 50271,
88
+ "<|nn|>": 50342,
89
+ "<|no|>": 50288,
90
+ "<|oc|>": 50328,
91
+ "<|pa|>": 50321,
92
+ "<|pl|>": 50269,
93
+ "<|ps|>": 50340,
94
+ "<|pt|>": 50267,
95
+ "<|ro|>": 50284,
96
+ "<|ru|>": 50263,
97
+ "<|sa|>": 50344,
98
+ "<|sd|>": 50332,
99
+ "<|si|>": 50322,
100
+ "<|sk|>": 50298,
101
+ "<|sl|>": 50305,
102
+ "<|sn|>": 50324,
103
+ "<|so|>": 50326,
104
+ "<|sq|>": 50317,
105
+ "<|sr|>": 50303,
106
+ "<|su|>": 50357,
107
+ "<|sv|>": 50273,
108
+ "<|sw|>": 50318,
109
+ "<|ta|>": 50287,
110
+ "<|te|>": 50299,
111
+ "<|tg|>": 50331,
112
+ "<|th|>": 50289,
113
+ "<|tk|>": 50341,
114
+ "<|tl|>": 50348,
115
+ "<|tr|>": 50268,
116
+ "<|tt|>": 50351,
117
+ "<|uk|>": 50280,
118
+ "<|ur|>": 50290,
119
+ "<|uz|>": 50337,
120
+ "<|vi|>": 50278,
121
+ "<|yi|>": 50335,
122
+ "<|yo|>": 50325,
123
+ "<|zh|>": 50260
124
+ },
125
+ "max_initial_timestamp_index": 1,
126
+ "max_length": 448,
127
+ "no_timestamps_token_id": 50363,
128
+ "pad_token_id": 50257,
129
+ "return_timestamps": false,
130
+ "suppress_tokens": [
131
+ 1,
132
+ 2,
133
+ 7,
134
+ 8,
135
+ 9,
136
+ 10,
137
+ 14,
138
+ 25,
139
+ 26,
140
+ 27,
141
+ 28,
142
+ 29,
143
+ 31,
144
+ 58,
145
+ 59,
146
+ 60,
147
+ 61,
148
+ 62,
149
+ 63,
150
+ 90,
151
+ 91,
152
+ 92,
153
+ 93,
154
+ 359,
155
+ 503,
156
+ 522,
157
+ 542,
158
+ 873,
159
+ 893,
160
+ 902,
161
+ 918,
162
+ 922,
163
+ 931,
164
+ 1350,
165
+ 1853,
166
+ 1982,
167
+ 2460,
168
+ 2627,
169
+ 3246,
170
+ 3253,
171
+ 3268,
172
+ 3536,
173
+ 3846,
174
+ 3961,
175
+ 4183,
176
+ 4667,
177
+ 6585,
178
+ 6647,
179
+ 7273,
180
+ 9061,
181
+ 9383,
182
+ 10428,
183
+ 10929,
184
+ 11938,
185
+ 12033,
186
+ 12331,
187
+ 12562,
188
+ 13793,
189
+ 14157,
190
+ 14635,
191
+ 15265,
192
+ 15618,
193
+ 16553,
194
+ 16604,
195
+ 18362,
196
+ 18956,
197
+ 20075,
198
+ 21675,
199
+ 22520,
200
+ 26130,
201
+ 26161,
202
+ 26435,
203
+ 28279,
204
+ 29464,
205
+ 31650,
206
+ 32302,
207
+ 32470,
208
+ 36865,
209
+ 42863,
210
+ 47425,
211
+ 49870,
212
+ 50254,
213
+ 50258,
214
+ 50358,
215
+ 50359,
216
+ 50360,
217
+ 50361,
218
+ 50362
219
+ ],
220
+ "task_to_id": {
221
+ "transcribe": 50359,
222
+ "translate": 50358
223
+ },
224
+ "transformers_version": "4.27.4"
225
+ }
preprocessor_config.json ADDED
The diff for this file is too large to render. See raw diff
 
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7bfbc41d27a9e7bfd1502fc4bd3081493012f0c6f48d074aca745ed882dcb371
3
+ size 3055741593
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": false,
3
+ "add_prefix_space": false,
4
+ "bos_token": {
5
+ "__type": "AddedToken",
6
+ "content": "<|endoftext|>",
7
+ "lstrip": false,
8
+ "normalized": true,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "eos_token": {
13
+ "__type": "AddedToken",
14
+ "content": "<|endoftext|>",
15
+ "lstrip": false,
16
+ "normalized": true,
17
+ "rstrip": false,
18
+ "single_word": false
19
+ },
20
+ "errors": "replace",
21
+ "model_max_length": 1024,
22
+ "name_or_path": "openai/whisper-medium",
23
+ "pad_token": null,
24
+ "processor_class": "WhisperProcessor",
25
+ "return_attention_mask": false,
26
+ "special_tokens_map_file": null,
27
+ "tokenizer_class": "WhisperTokenizer",
28
+ "unk_token": {
29
+ "__type": "AddedToken",
30
+ "content": "<|endoftext|>",
31
+ "lstrip": false,
32
+ "normalized": true,
33
+ "rstrip": false,
34
+ "single_word": false
35
+ }
36
+ }
vocab.json ADDED
The diff for this file is too large to render. See raw diff