zary0 commited on
Commit
dceb749
·
verified ·
1 Parent(s): 7a7e8db

Upload 3 files

Browse files
Files changed (3) hide show
  1. config.json +56 -199
  2. preprocessor_config.json +15 -0
  3. tokenizer.json +0 -0
config.json CHANGED
@@ -1,203 +1,60 @@
1
  {
2
- "alignment_heads": [
3
- [
4
- 3,
5
- 0
6
- ],
7
- [
8
- 3,
9
- 1
10
- ],
11
- [
12
- 3,
13
- 2
14
- ],
15
- [
16
- 3,
17
- 3
18
- ],
19
- [
20
- 3,
21
- 4
22
- ],
23
- [
24
- 3,
25
- 5
26
- ],
27
- [
28
- 3,
29
- 6
30
- ],
31
- [
32
- 3,
33
- 7
34
- ],
35
- [
36
- 4,
37
- 0
38
- ],
39
- [
40
- 4,
41
- 1
42
- ],
43
- [
44
- 4,
45
- 2
46
- ],
47
- [
48
- 4,
49
- 3
50
- ],
51
- [
52
- 4,
53
- 4
54
- ],
55
- [
56
- 4,
57
- 5
58
- ],
59
- [
60
- 4,
61
- 6
62
- ],
63
- [
64
- 4,
65
- 7
66
- ],
67
- [
68
- 5,
69
- 0
70
- ],
71
- [
72
- 5,
73
- 1
74
- ],
75
- [
76
- 5,
77
- 2
78
- ],
79
- [
80
- 5,
81
- 3
82
- ],
83
- [
84
- 5,
85
- 4
86
- ],
87
- [
88
- 5,
89
- 5
90
- ],
91
- [
92
- 5,
93
- 6
94
- ],
95
- [
96
- 5,
97
- 7
98
- ]
99
  ],
100
- "lang_ids": [
101
- 50259,
102
- 50260,
103
- 50261,
104
- 50262,
105
- 50263,
106
- 50264,
107
- 50265,
108
- 50266,
109
- 50267,
110
- 50268,
111
- 50269,
112
- 50270,
113
- 50271,
114
- 50272,
115
- 50273,
116
- 50274,
117
- 50275,
118
- 50276,
119
- 50277,
120
- 50278,
121
- 50279,
122
- 50280,
123
- 50281,
124
- 50282,
125
- 50283,
126
- 50284,
127
- 50285,
128
- 50286,
129
- 50287,
130
- 50288,
131
- 50289,
132
- 50290,
133
- 50291,
134
- 50292,
135
- 50293,
136
- 50294,
137
- 50295,
138
- 50296,
139
- 50297,
140
- 50298,
141
- 50299,
142
- 50300,
143
- 50301,
144
- 50302,
145
- 50303,
146
- 50304,
147
- 50305,
148
- 50306,
149
- 50307,
150
- 50308,
151
- 50309,
152
- 50310,
153
- 50311,
154
- 50312,
155
- 50313,
156
- 50314,
157
- 50315,
158
- 50316,
159
- 50317,
160
- 50318,
161
- 50319,
162
- 50320,
163
- 50321,
164
- 50322,
165
- 50323,
166
- 50324,
167
- 50325,
168
- 50326,
169
- 50327,
170
- 50328,
171
- 50329,
172
- 50330,
173
- 50331,
174
- 50332,
175
- 50333,
176
- 50334,
177
- 50335,
178
- 50336,
179
- 50337,
180
- 50338,
181
- 50339,
182
- 50340,
183
- 50341,
184
- 50342,
185
- 50343,
186
- 50344,
187
- 50345,
188
- 50346,
189
- 50347,
190
- 50348,
191
- 50349,
192
- 50350,
193
- 50351,
194
- 50352,
195
- 50353,
196
- 50354,
197
- 50355,
198
- 50356,
199
- 50357
200
  ],
201
- "suppress_ids": null,
202
- "suppress_ids_begin": null
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
203
  }
 
1
  {
2
+ "activation_dropout": 0.0,
3
+ "activation_function": "gelu",
4
+ "apply_spec_augment": false,
5
+ "architectures": [
6
+ "WhisperForConditionalGeneration"
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
7
  ],
8
+ "attention_dropout": 0.0,
9
+ "begin_suppress_tokens": null,
10
+ "bos_token_id": 50257,
11
+ "classifier_proj_size": 256,
12
+ "d_model": 512,
13
+ "decoder_attention_heads": 8,
14
+ "decoder_ffn_dim": 2048,
15
+ "decoder_layerdrop": 0.0,
16
+ "decoder_layers": 6,
17
+ "decoder_start_token_id": 50258,
18
+ "dropout": 0.0,
19
+ "dtype": "float32",
20
+ "encoder_attention_heads": 8,
21
+ "encoder_ffn_dim": 2048,
22
+ "encoder_layerdrop": 0.0,
23
+ "encoder_layers": 6,
24
+ "eos_token_id": 50257,
25
+ "forced_decoder_ids": [
26
+ [
27
+ 1,
28
+ 50266
29
+ ],
30
+ [
31
+ 2,
32
+ 50359
33
+ ],
34
+ [
35
+ 3,
36
+ 50363
37
+ ]
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
38
  ],
39
+ "init_std": 0.02,
40
+ "is_encoder_decoder": true,
41
+ "mask_feature_length": 10,
42
+ "mask_feature_min_masks": 0,
43
+ "mask_feature_prob": 0.0,
44
+ "mask_time_length": 10,
45
+ "mask_time_min_masks": 2,
46
+ "mask_time_prob": 0.05,
47
+ "max_length": null,
48
+ "max_source_positions": 1500,
49
+ "max_target_positions": 448,
50
+ "median_filter_width": 7,
51
+ "model_type": "whisper",
52
+ "num_hidden_layers": 6,
53
+ "num_mel_bins": 80,
54
+ "pad_token_id": 50257,
55
+ "scale_embedding": false,
56
+ "transformers_version": "4.56.1",
57
+ "use_cache": true,
58
+ "use_weighted_layer_sum": false,
59
+ "vocab_size": 51865
60
  }
preprocessor_config.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "chunk_length": 30,
3
+ "dither": 0.0,
4
+ "feature_extractor_type": "WhisperFeatureExtractor",
5
+ "feature_size": 80,
6
+ "hop_length": 160,
7
+ "n_fft": 400,
8
+ "n_samples": 480000,
9
+ "nb_max_frames": 3000,
10
+ "padding_side": "right",
11
+ "padding_value": 0.0,
12
+ "processor_class": "WhisperProcessor",
13
+ "return_attention_mask": false,
14
+ "sampling_rate": 16000
15
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff