Hugging Face's logo

Aanchan
/
psst_model_cer_2

Automatic Speech Recognition
Transformers
PyTorch
wav2vec2
Generated from Trainer
Model card Files
xet
 Community
4
Aanchan commited on
Commit
645f031
·
1 Parent(s): 965c5fc

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +293 -0
README.md ADDED
@@ -0,0 +1,293 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ tags:
4
+ - generated_from_trainer
5
+ model-index:
6
+ - name: psst_model_cer_2
7
+ results: []
8
+ ---
9
+
10
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
11
+ should probably proofread and complete it, then remove this comment. -->
12
+
13
+ # psst_model_cer_2
14
+
15
+ This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the None dataset.
16
+ It achieves the following results on the evaluation set:
17
+ - Loss: 2.0086
18
+ - Cer: 0.5952
19
+
20
+ ## Model description
21
+
22
+ More information needed
23
+
24
+ ## Intended uses & limitations
25
+
26
+ More information needed
27
+
28
+ ## Training and evaluation data
29
+
30
+ More information needed
31
+
32
+ ## Training procedure
33
+
34
+ ### Training hyperparameters
35
+
36
+ The following hyperparameters were used during training:
37
+ - learning_rate: 5e-05
38
+ - train_batch_size: 6
39
+ - eval_batch_size: 6
40
+ - seed: 42
41
+ - gradient_accumulation_steps: 8
42
+ - total_train_batch_size: 48
43
+ - optimizer: Adam with betas=(0.9,0.98) and epsilon=1e-08
44
+ - lr_scheduler_type: linear
45
+ - num_epochs: 50
46
+
47
+ ### Training results
48
+
49
+ | Training Loss | Epoch | Step | Validation Loss | Cer |
50
+ |:-------------:|:-----:|:----:|:---------------:|:------:|
51
+ | 37.8359 | 0.21 | 10 | 33.3633 | 1.0 |
52
+ | 10.3672 | 0.42 | 20 | 12.7370 | 1.0 |
53
+ | 6.965 | 0.63 | 30 | 10.1163 | 1.0 |
54
+ | 5.4782 | 0.84 | 40 | 8.5443 | 1.0 |
55
+ | 4.9771 | 1.04 | 50 | 7.6301 | 1.0 |
56
+ | 4.4938 | 1.25 | 60 | 6.9782 | 1.0 |
57
+ | 4.201 | 1.46 | 70 | 6.3865 | 1.0 |
58
+ | 4.268 | 1.67 | 80 | 6.3360 | 1.0 |
59
+ | 4.146 | 1.88 | 90 | 5.7938 | 1.0 |
60
+ | 4.1606 | 2.09 | 100 | 5.6862 | 1.0 |
61
+ | 4.0541 | 2.3 | 110 | 5.6427 | 1.0 |
62
+ | 4.0327 | 2.51 | 120 | 6.0837 | 1.0 |
63
+ | 3.9125 | 2.72 | 130 | 5.6241 | 1.0 |
64
+ | 3.9092 | 2.92 | 140 | 5.3858 | 1.0 |
65
+ | 3.9797 | 3.13 | 150 | 5.0597 | 1.0 |
66
+ | 3.909 | 3.34 | 160 | 5.2240 | 1.0 |
67
+ | 3.9256 | 3.55 | 170 | 5.1131 | 1.0 |
68
+ | 3.8817 | 3.76 | 180 | 5.3508 | 1.0 |
69
+ | 3.8973 | 3.97 | 190 | 5.6252 | 1.0 |
70
+ | 3.9605 | 4.18 | 200 | 5.2654 | 1.0 |
71
+ | 3.91 | 4.39 | 210 | 5.2997 | 1.0 |
72
+ | 3.9179 | 4.6 | 220 | 5.1039 | 1.0 |
73
+ | 3.887 | 4.8 | 230 | 5.1672 | 1.0 |
74
+ | 3.8978 | 5.01 | 240 | 5.2943 | 1.0 |
75
+ | 3.8155 | 5.22 | 250 | 5.5301 | 1.0 |
76
+ | 3.9853 | 5.43 | 260 | 5.1155 | 1.0 |
77
+ | 3.8906 | 5.64 | 270 | 5.0986 | 1.0 |
78
+ | 3.8601 | 5.85 | 280 | 4.6896 | 1.0 |
79
+ | 3.9491 | 6.06 | 290 | 5.2097 | 1.0 |
80
+ | 3.8721 | 6.27 | 300 | 4.8001 | 1.0 |
81
+ | 3.8459 | 6.48 | 310 | 4.8305 | 1.0 |
82
+ | 3.8119 | 6.68 | 320 | 5.4870 | 1.0 |
83
+ | 3.8184 | 6.89 | 330 | 4.5510 | 1.0 |
84
+ | 3.8064 | 7.1 | 340 | 5.0332 | 1.0 |
85
+ | 3.8516 | 7.31 | 350 | 4.6039 | 1.0 |
86
+ | 3.8689 | 7.52 | 360 | 4.9772 | 1.0 |
87
+ | 3.7968 | 7.73 | 370 | 4.4847 | 1.0 |
88
+ | 3.8069 | 7.94 | 380 | 4.5842 | 1.0 |
89
+ | 3.7918 | 8.15 | 390 | 4.4859 | 1.0 |
90
+ | 3.7672 | 8.36 | 400 | 4.2332 | 1.0 |
91
+ | 3.7639 | 8.56 | 410 | 4.2703 | 1.0 |
92
+ | 3.8094 | 8.77 | 420 | 4.5038 | 1.0 |
93
+ | 3.7718 | 8.98 | 430 | 4.2372 | 1.0 |
94
+ | 3.7588 | 9.19 | 440 | 4.1989 | 1.0 |
95
+ | 3.7504 | 9.4 | 450 | 4.2191 | 1.0 |
96
+ | 3.8373 | 9.61 | 460 | 4.1814 | 1.0 |
97
+ | 3.7472 | 9.82 | 470 | 4.2161 | 1.0 |
98
+ | 3.7998 | 10.03 | 480 | 4.1128 | 1.0 |
99
+ | 3.7542 | 10.23 | 490 | 4.2112 | 1.0 |
100
+ | 3.7366 | 10.44 | 500 | 4.1716 | 1.0 |
101
+ | 3.7243 | 10.65 | 510 | 4.0599 | 1.0 |
102
+ | 3.7427 | 10.86 | 520 | 4.0381 | 1.0 |
103
+ | 3.7373 | 11.07 | 530 | 4.0331 | 1.0 |
104
+ | 3.7367 | 11.28 | 540 | 4.0185 | 1.0 |
105
+ | 3.702 | 11.49 | 550 | 3.9956 | 1.0 |
106
+ | 3.7854 | 11.7 | 560 | 4.0306 | 1.0 |
107
+ | 3.741 | 11.91 | 570 | 4.1382 | 1.0 |
108
+ | 3.6995 | 12.11 | 580 | 4.0207 | 1.0 |
109
+ | 3.7248 | 12.32 | 590 | 4.0219 | 1.0 |
110
+ | 3.7232 | 12.53 | 600 | 4.0050 | 1.0 |
111
+ | 3.7246 | 12.74 | 610 | 4.0153 | 1.0 |
112
+ | 3.7134 | 12.95 | 620 | 4.0040 | 1.0 |
113
+ | 3.6979 | 13.16 | 630 | 4.0014 | 1.0 |
114
+ | 3.7051 | 13.37 | 640 | 4.0137 | 1.0 |
115
+ | 3.6915 | 13.58 | 650 | 3.9692 | 1.0 |
116
+ | 3.7047 | 13.79 | 660 | 3.9598 | 1.0 |
117
+ | 3.7833 | 13.99 | 670 | 3.9548 | 1.0 |
118
+ | 3.7084 | 14.2 | 680 | 3.9840 | 1.0 |
119
+ | 3.7876 | 14.41 | 690 | 4.0253 | 1.0 |
120
+ | 3.7278 | 14.62 | 700 | 3.9687 | 1.0 |
121
+ | 3.7019 | 14.83 | 710 | 3.9461 | 1.0 |
122
+ | 3.6867 | 15.04 | 720 | 3.8824 | 1.0 |
123
+ | 3.6685 | 15.25 | 730 | 3.9191 | 1.0 |
124
+ | 3.6785 | 15.46 | 740 | 3.8702 | 1.0 |
125
+ | 3.6193 | 15.67 | 750 | 4.0987 | 1.0 |
126
+ | 3.6695 | 15.87 | 760 | 3.8417 | 1.0 |
127
+ | 3.6322 | 16.08 | 770 | 3.8697 | 1.0 |
128
+ | 3.6423 | 16.29 | 780 | 4.0140 | 1.0 |
129
+ | 3.6224 | 16.5 | 790 | 3.8780 | 1.0 |
130
+ | 3.5836 | 16.71 | 800 | 3.8246 | 1.0 |
131
+ | 3.6403 | 16.92 | 810 | 3.8050 | 1.0 |
132
+ | 3.5728 | 17.13 | 820 | 3.7491 | 1.0 |
133
+ | 3.5611 | 17.34 | 830 | 3.7878 | 1.0 |
134
+ | 3.5573 | 17.55 | 840 | 3.7570 | 1.0 |
135
+ | 3.5202 | 17.75 | 850 | 3.7969 | 1.0 |
136
+ | 3.5196 | 17.96 | 860 | 3.6973 | 1.0 |
137
+ | 3.5235 | 18.17 | 870 | 3.6411 | 1.0 |
138
+ | 3.4898 | 18.38 | 880 | 3.7192 | 1.0 |
139
+ | 3.5062 | 18.59 | 890 | 3.6198 | 1.0 |
140
+ | 3.4717 | 18.8 | 900 | 3.6294 | 1.0 |
141
+ | 3.4377 | 19.01 | 910 | 3.6018 | 1.0 |
142
+ | 3.4012 | 19.22 | 920 | 3.6305 | 1.0 |
143
+ | 3.4097 | 19.43 | 930 | 3.5503 | 1.0 |
144
+ | 3.4805 | 19.63 | 940 | 3.5701 | 1.0 |
145
+ | 3.4145 | 19.84 | 950 | 3.5523 | 1.0 |
146
+ | 3.4065 | 20.05 | 960 | 3.5159 | 1.0 |
147
+ | 3.3682 | 20.26 | 970 | 3.4709 | 1.0 |
148
+ | 3.342 | 20.47 | 980 | 3.4900 | 1.0 |
149
+ | 3.3573 | 20.68 | 990 | 3.6064 | 0.9999 |
150
+ | 3.3733 | 20.89 | 1000 | 3.4031 | 1.0 |
151
+ | 3.317 | 21.1 | 1010 | 3.4064 | 1.0 |
152
+ | 3.3028 | 21.31 | 1020 | 3.3974 | 0.9995 |
153
+ | 3.2726 | 21.51 | 1030 | 3.4504 | 0.9996 |
154
+ | 3.2676 | 21.72 | 1040 | 3.3540 | 0.9996 |
155
+ | 3.2661 | 21.93 | 1050 | 3.3113 | 0.9997 |
156
+ | 3.2838 | 22.14 | 1060 | 3.3126 | 0.9999 |
157
+ | 3.203 | 22.35 | 1070 | 3.2989 | 0.9967 |
158
+ | 3.2241 | 22.56 | 1080 | 3.3561 | 0.9927 |
159
+ | 3.2106 | 22.77 | 1090 | 3.2749 | 0.9944 |
160
+ | 3.2173 | 22.98 | 1100 | 3.1937 | 0.9989 |
161
+ | 3.2015 | 23.19 | 1110 | 3.2013 | 0.9989 |
162
+ | 3.1684 | 23.39 | 1120 | 3.1966 | 0.9981 |
163
+ | 3.1692 | 23.6 | 1130 | 3.1515 | 0.9903 |
164
+ | 3.153 | 23.81 | 1140 | 3.1713 | 0.9658 |
165
+ | 3.1614 | 24.02 | 1150 | 3.0817 | 0.9883 |
166
+ | 3.1328 | 24.23 | 1160 | 3.1405 | 0.9879 |
167
+ | 3.1528 | 24.44 | 1170 | 3.0858 | 0.9586 |
168
+ | 3.1284 | 24.65 | 1180 | 3.1450 | 0.9408 |
169
+ | 3.161 | 24.86 | 1190 | 3.0577 | 0.9772 |
170
+ | 3.1112 | 25.07 | 1200 | 3.0405 | 0.9788 |
171
+ | 3.1134 | 25.27 | 1210 | 3.0256 | 0.9388 |
172
+ | 3.0451 | 25.48 | 1220 | 3.0475 | 0.9388 |
173
+ | 3.0835 | 25.69 | 1230 | 3.0259 | 0.9912 |
174
+ | 3.151 | 25.9 | 1240 | 2.9855 | 0.9808 |
175
+ | 3.1043 | 26.11 | 1250 | 2.9574 | 0.9548 |
176
+ | 3.0278 | 26.32 | 1260 | 2.9594 | 0.9338 |
177
+ | 3.0517 | 26.53 | 1270 | 2.9066 | 0.9029 |
178
+ | 2.9938 | 26.74 | 1280 | 2.8813 | 0.9042 |
179
+ | 2.9979 | 26.95 | 1290 | 2.8485 | 0.9084 |
180
+ | 3.0262 | 27.15 | 1300 | 2.8382 | 0.8840 |
181
+ | 3.0048 | 27.36 | 1310 | 2.8424 | 0.8933 |
182
+ | 2.9676 | 27.57 | 1320 | 2.8168 | 0.8861 |
183
+ | 2.972 | 27.78 | 1330 | 2.8324 | 0.8949 |
184
+ | 2.9057 | 27.99 | 1340 | 2.7582 | 0.8645 |
185
+ | 2.9145 | 28.2 | 1350 | 2.7261 | 0.8555 |
186
+ | 2.8708 | 28.41 | 1360 | 2.7156 | 0.8230 |
187
+ | 2.9127 | 28.62 | 1370 | 2.7340 | 0.8738 |
188
+ | 2.9277 | 28.83 | 1380 | 2.7288 | 0.8539 |
189
+ | 2.8918 | 29.03 | 1390 | 2.6852 | 0.8083 |
190
+ | 2.9109 | 29.24 | 1400 | 2.6678 | 0.8243 |
191
+ | 2.8655 | 29.45 | 1410 | 2.7691 | 0.8872 |
192
+ | 2.9382 | 29.66 | 1420 | 2.6778 | 0.8350 |
193
+ | 2.8375 | 29.87 | 1430 | 2.6705 | 0.7697 |
194
+ | 2.8531 | 30.08 | 1440 | 2.6343 | 0.8125 |
195
+ | 2.8456 | 30.29 | 1450 | 2.6113 | 0.7724 |
196
+ | 2.8201 | 30.5 | 1460 | 2.5863 | 0.7613 |
197
+ | 2.8172 | 30.7 | 1470 | 2.6602 | 0.8386 |
198
+ | 2.8465 | 30.91 | 1480 | 2.5719 | 0.8038 |
199
+ | 2.8527 | 31.12 | 1490 | 2.5469 | 0.7544 |
200
+ | 2.7747 | 31.33 | 1500 | 2.5257 | 0.7847 |
201
+ | 2.7948 | 31.54 | 1510 | 2.5263 | 0.7909 |
202
+ | 2.6885 | 31.75 | 1520 | 2.4958 | 0.7364 |
203
+ | 2.7594 | 31.96 | 1530 | 2.5140 | 0.7786 |
204
+ | 2.7385 | 32.17 | 1540 | 2.4742 | 0.7775 |
205
+ | 2.8526 | 32.38 | 1550 | 2.4751 | 0.6944 |
206
+ | 2.7398 | 32.58 | 1560 | 2.4276 | 0.7372 |
207
+ | 2.8246 | 32.79 | 1570 | 2.4695 | 0.7523 |
208
+ | 2.7004 | 33.0 | 1580 | 2.4228 | 0.7059 |
209
+ | 2.7431 | 33.21 | 1590 | 2.4395 | 0.7569 |
210
+ | 2.7673 | 33.42 | 1600 | 2.4158 | 0.7011 |
211
+ | 2.7558 | 33.63 | 1610 | 2.3673 | 0.6953 |
212
+ | 2.7892 | 33.84 | 1620 | 2.3885 | 0.7284 |
213
+ | 2.7038 | 34.05 | 1630 | 2.3655 | 0.7089 |
214
+ | 2.6997 | 34.26 | 1640 | 2.3426 | 0.7120 |
215
+ | 2.6738 | 34.46 | 1650 | 2.3455 | 0.7335 |
216
+ | 2.6899 | 34.67 | 1660 | 2.3281 | 0.6771 |
217
+ | 2.6903 | 34.88 | 1670 | 2.3271 | 0.6825 |
218
+ | 2.6737 | 35.09 | 1680 | 2.3032 | 0.6826 |
219
+ | 2.5978 | 35.3 | 1690 | 2.2940 | 0.7169 |
220
+ | 2.6285 | 35.51 | 1700 | 2.3023 | 0.6771 |
221
+ | 2.7255 | 35.72 | 1710 | 2.3156 | 0.6726 |
222
+ | 2.5993 | 35.93 | 1720 | 2.3340 | 0.7492 |
223
+ | 2.6732 | 36.14 | 1730 | 2.2575 | 0.6537 |
224
+ | 2.6294 | 36.34 | 1740 | 2.2502 | 0.6561 |
225
+ | 2.6515 | 36.55 | 1750 | 2.2356 | 0.6746 |
226
+ | 2.5892 | 36.76 | 1760 | 2.2364 | 0.6567 |
227
+ | 2.6174 | 36.97 | 1770 | 2.2375 | 0.6655 |
228
+ | 2.6555 | 37.18 | 1780 | 2.2259 | 0.6861 |
229
+ | 2.591 | 37.39 | 1790 | 2.2053 | 0.6489 |
230
+ | 2.5531 | 37.6 | 1800 | 2.2058 | 0.6295 |
231
+ | 2.5632 | 37.81 | 1810 | 2.2065 | 0.6719 |
232
+ | 2.5021 | 38.02 | 1820 | 2.1740 | 0.6371 |
233
+ | 2.5258 | 38.22 | 1830 | 2.1784 | 0.6515 |
234
+ | 2.5817 | 38.43 | 1840 | 2.1813 | 0.6749 |
235
+ | 2.6251 | 38.64 | 1850 | 2.1902 | 0.6165 |
236
+ | 2.6398 | 38.85 | 1860 | 2.1529 | 0.6557 |
237
+ | 2.5636 | 39.06 | 1870 | 2.1458 | 0.6429 |
238
+ | 2.5731 | 39.27 | 1880 | 2.1535 | 0.6435 |
239
+ | 2.5461 | 39.48 | 1890 | 2.1516 | 0.6380 |
240
+ | 2.5975 | 39.69 | 1900 | 2.1540 | 0.6557 |
241
+ | 2.5979 | 39.9 | 1910 | 2.1437 | 0.6389 |
242
+ | 2.5359 | 40.1 | 1920 | 2.1334 | 0.6201 |
243
+ | 2.5004 | 40.31 | 1930 | 2.1152 | 0.6371 |
244
+ | 2.5928 | 40.52 | 1940 | 2.1069 | 0.6079 |
245
+ | 2.5325 | 40.73 | 1950 | 2.1154 | 0.6341 |
246
+ | 2.5149 | 40.94 | 1960 | 2.1152 | 0.6131 |
247
+ | 2.5635 | 41.15 | 1970 | 2.1097 | 0.6298 |
248
+ | 2.6024 | 41.36 | 1980 | 2.1066 | 0.6384 |
249
+ | 2.5074 | 41.57 | 1990 | 2.1147 | 0.5998 |
250
+ | 2.5191 | 41.78 | 2000 | 2.0933 | 0.6166 |
251
+ | 2.4896 | 41.98 | 2010 | 2.0805 | 0.6063 |
252
+ | 2.4983 | 42.19 | 2020 | 2.0860 | 0.6189 |
253
+ | 2.5139 | 42.4 | 2030 | 2.0893 | 0.6064 |
254
+ | 2.4801 | 42.61 | 2040 | 2.0701 | 0.6176 |
255
+ | 2.541 | 42.82 | 2050 | 2.0674 | 0.6122 |
256
+ | 2.4728 | 43.03 | 2060 | 2.0571 | 0.6123 |
257
+ | 2.4672 | 43.24 | 2070 | 2.0522 | 0.6128 |
258
+ | 2.4716 | 43.45 | 2080 | 2.0559 | 0.5965 |
259
+ | 2.4621 | 43.66 | 2090 | 2.0512 | 0.6050 |
260
+ | 2.4838 | 43.86 | 2100 | 2.0431 | 0.6057 |
261
+ | 2.512 | 44.07 | 2110 | 2.0378 | 0.5995 |
262
+ | 2.4825 | 44.28 | 2120 | 2.0394 | 0.6075 |
263
+ | 2.4814 | 44.49 | 2130 | 2.0368 | 0.5973 |
264
+ | 2.4023 | 44.7 | 2140 | 2.0482 | 0.6219 |
265
+ | 2.494 | 44.91 | 2150 | 2.0611 | 0.5765 |
266
+ | 2.5854 | 45.12 | 2160 | 2.0383 | 0.6125 |
267
+ | 2.4052 | 45.33 | 2170 | 2.0348 | 0.5935 |
268
+ | 2.4368 | 45.54 | 2180 | 2.0323 | 0.6002 |
269
+ | 2.3926 | 45.74 | 2190 | 2.0292 | 0.6107 |
270
+ | 2.4397 | 45.95 | 2200 | 2.0335 | 0.5873 |
271
+ | 2.5227 | 46.16 | 2210 | 2.0242 | 0.6075 |
272
+ | 2.4675 | 46.37 | 2220 | 2.0237 | 0.5992 |
273
+ | 2.4815 | 46.58 | 2230 | 2.0222 | 0.5956 |
274
+ | 2.4211 | 46.79 | 2240 | 2.0187 | 0.5953 |
275
+ | 2.3833 | 47.0 | 2250 | 2.0173 | 0.5898 |
276
+ | 2.4827 | 47.21 | 2260 | 2.0137 | 0.5993 |
277
+ | 2.5543 | 47.42 | 2270 | 2.0131 | 0.5977 |
278
+ | 2.4751 | 47.62 | 2280 | 2.0117 | 0.5956 |
279
+ | 2.4562 | 47.83 | 2290 | 2.0124 | 0.5979 |
280
+ | 2.4648 | 48.04 | 2300 | 2.0127 | 0.5992 |
281
+ | 2.4327 | 48.25 | 2310 | 2.0121 | 0.5930 |
282
+ | 2.4454 | 48.46 | 2320 | 2.0098 | 0.5969 |
283
+ | 2.4821 | 48.67 | 2330 | 2.0088 | 0.5962 |
284
+ | 2.4527 | 48.88 | 2340 | 2.0087 | 0.5953 |
285
+ | 2.4899 | 49.09 | 2350 | 2.0086 | 0.5952 |
286
+
287
+
288
+ ### Framework versions
289
+
290
+ - Transformers 4.28.0
291
+ - Pytorch 2.0.1+cu117
292
+ - Datasets 2.14.4
293
+ - Tokenizers 0.13.3