ngkien commited on
Commit
80ddfce
1 Parent(s): 1402fac

Upload the Distil-Whisper medium.en conversion model to the Hugging Face hub

Browse files
Files changed (7) hide show
  1. .gitattributes +1 -0
  2. README.md +39 -0
  3. config.json +265 -0
  4. model.bin +3 -0
  5. preprocessor_config.json +14 -0
  6. tokenizer.json +0 -0
  7. vocabulary.json +0 -0
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ model.bin filter=lfs diff=lfs merge=lfs -text
README.md CHANGED
@@ -1,3 +1,42 @@
1
  ---
 
 
 
 
 
2
  license: mit
 
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ language:
3
+ - en
4
+ tags:
5
+ - audio
6
+ - automatic-speech-recognition
7
  license: mit
8
+ library_name: ctranslate2
9
  ---
10
+
11
+ # Whisper medium.en model for CTranslate2
12
+
13
+ This repository contains the conversion of [distil-whisper/distil-medium.en](https://huggingface.co/distil-whisper/distil-medium.en) to the [CTranslate2](https://github.com/OpenNMT/CTranslate2) model format.
14
+
15
+ This model can be used in CTranslate2 or projects based on CTranslate2 such as [faster-whisper](https://github.com/systran/faster-whisper).
16
+
17
+ ## Example
18
+
19
+ ```python
20
+ from faster_whisper import WhisperModel
21
+
22
+ model = WhisperModel("distil-medium.en")
23
+
24
+ segments, info = model.transcribe("audio.mp3")
25
+ for segment in segments:
26
+ print("[%.2fs -> %.2fs] %s" % (segment.start, segment.end, segment.text))
27
+ ```
28
+
29
+ ## Conversion details
30
+
31
+ The original model was converted with the following command:
32
+
33
+ ```
34
+ ct2-transformers-converter --model distil-whisper/distil-medium.en --output_dir faster-distil-whisper-medium.en \
35
+ --copy_files tokenizer.json preprocessor_config.json --quantization float16
36
+ ```
37
+
38
+ Note that the model weights are saved in FP16. This type can be changed when the model is loaded using the [`compute_type` option in CTranslate2](https://opennmt.net/CTranslate2/quantization.html).
39
+
40
+ ## More information
41
+
42
+ **For more information about the original model, see its [model card](https://huggingface.co/distil-whisper/distil-medium.en).**
config.json ADDED
@@ -0,0 +1,265 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alignment_heads": [
3
+ [
4
+ 1,
5
+ 0
6
+ ],
7
+ [
8
+ 1,
9
+ 1
10
+ ],
11
+ [
12
+ 1,
13
+ 2
14
+ ],
15
+ [
16
+ 1,
17
+ 3
18
+ ],
19
+ [
20
+ 1,
21
+ 4
22
+ ],
23
+ [
24
+ 1,
25
+ 5
26
+ ],
27
+ [
28
+ 1,
29
+ 6
30
+ ],
31
+ [
32
+ 1,
33
+ 7
34
+ ],
35
+ [
36
+ 1,
37
+ 8
38
+ ],
39
+ [
40
+ 1,
41
+ 9
42
+ ],
43
+ [
44
+ 1,
45
+ 10
46
+ ],
47
+ [
48
+ 1,
49
+ 11
50
+ ],
51
+ [
52
+ 1,
53
+ 12
54
+ ],
55
+ [
56
+ 1,
57
+ 13
58
+ ],
59
+ [
60
+ 1,
61
+ 14
62
+ ],
63
+ [
64
+ 1,
65
+ 15
66
+ ]
67
+ ],
68
+ "lang_ids": [
69
+ 50258,
70
+ 50259,
71
+ 50260,
72
+ 50261,
73
+ 50262,
74
+ 50263,
75
+ 50264,
76
+ 50265,
77
+ 50266,
78
+ 50267,
79
+ 50268,
80
+ 50269,
81
+ 50270,
82
+ 50271,
83
+ 50272,
84
+ 50273,
85
+ 50274,
86
+ 50275,
87
+ 50276,
88
+ 50277,
89
+ 50278,
90
+ 50279,
91
+ 50280,
92
+ 50281,
93
+ 50282,
94
+ 50283,
95
+ 50284,
96
+ 50285,
97
+ 50286,
98
+ 50287,
99
+ 50288,
100
+ 50289,
101
+ 50290,
102
+ 50291,
103
+ 50292,
104
+ 50293,
105
+ 50294,
106
+ 50295,
107
+ 50296,
108
+ 50297,
109
+ 50298,
110
+ 50299,
111
+ 50300,
112
+ 50301,
113
+ 50302,
114
+ 50303,
115
+ 50304,
116
+ 50305,
117
+ 50306,
118
+ 50307,
119
+ 50308,
120
+ 50309,
121
+ 50310,
122
+ 50311,
123
+ 50312,
124
+ 50313,
125
+ 50314,
126
+ 50315,
127
+ 50316,
128
+ 50317,
129
+ 50318,
130
+ 50319,
131
+ 50320,
132
+ 50321,
133
+ 50322,
134
+ 50323,
135
+ 50324,
136
+ 50325,
137
+ 50326,
138
+ 50327,
139
+ 50328,
140
+ 50329,
141
+ 50330,
142
+ 50331,
143
+ 50332,
144
+ 50333,
145
+ 50334,
146
+ 50335,
147
+ 50336,
148
+ 50337,
149
+ 50338,
150
+ 50339,
151
+ 50340,
152
+ 50341,
153
+ 50342,
154
+ 50343,
155
+ 50344,
156
+ 50345,
157
+ 50346,
158
+ 50347,
159
+ 50348,
160
+ 50349,
161
+ 50350,
162
+ 50351,
163
+ 50352,
164
+ 50353,
165
+ 50354,
166
+ 50355,
167
+ 50356
168
+ ],
169
+ "suppress_ids": [
170
+ 1,
171
+ 2,
172
+ 7,
173
+ 8,
174
+ 9,
175
+ 10,
176
+ 14,
177
+ 25,
178
+ 26,
179
+ 27,
180
+ 28,
181
+ 29,
182
+ 31,
183
+ 58,
184
+ 59,
185
+ 60,
186
+ 61,
187
+ 62,
188
+ 63,
189
+ 90,
190
+ 91,
191
+ 92,
192
+ 93,
193
+ 357,
194
+ 366,
195
+ 438,
196
+ 532,
197
+ 685,
198
+ 705,
199
+ 796,
200
+ 930,
201
+ 1058,
202
+ 1220,
203
+ 1267,
204
+ 1279,
205
+ 1303,
206
+ 1343,
207
+ 1377,
208
+ 1391,
209
+ 1635,
210
+ 1782,
211
+ 1875,
212
+ 2162,
213
+ 2361,
214
+ 2488,
215
+ 3467,
216
+ 4008,
217
+ 4211,
218
+ 4600,
219
+ 4808,
220
+ 5299,
221
+ 5855,
222
+ 6329,
223
+ 7203,
224
+ 9609,
225
+ 9959,
226
+ 10563,
227
+ 10786,
228
+ 11420,
229
+ 11709,
230
+ 11907,
231
+ 13163,
232
+ 13697,
233
+ 13700,
234
+ 14808,
235
+ 15306,
236
+ 16410,
237
+ 16791,
238
+ 17992,
239
+ 19203,
240
+ 19510,
241
+ 20724,
242
+ 22305,
243
+ 22935,
244
+ 27007,
245
+ 30109,
246
+ 30420,
247
+ 33409,
248
+ 34949,
249
+ 40283,
250
+ 40493,
251
+ 40549,
252
+ 47282,
253
+ 49146,
254
+ 50257,
255
+ 50357,
256
+ 50358,
257
+ 50359,
258
+ 50360,
259
+ 50361
260
+ ],
261
+ "suppress_ids_begin": [
262
+ 220,
263
+ 50256
264
+ ]
265
+ }
model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d4cb75d823dcd2647191064da76f026774c06c036908f38456165368d0e2d66a
3
+ size 788826555
preprocessor_config.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "chunk_length": 30,
3
+ "feature_extractor_type": "WhisperFeatureExtractor",
4
+ "feature_size": 80,
5
+ "hop_length": 160,
6
+ "n_fft": 400,
7
+ "n_samples": 480000,
8
+ "nb_max_frames": 3000,
9
+ "padding_side": "right",
10
+ "padding_value": 0.0,
11
+ "processor_class": "WhisperProcessor",
12
+ "return_attention_mask": false,
13
+ "sampling_rate": 16000
14
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
vocabulary.json ADDED
The diff for this file is too large to render. See raw diff