Whisper small vi - Ox
This model is a fine-tuned version of weights/whisper-small-vi on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.1208
- Wer: 22.6497
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 3.0
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Wer |
---|---|---|---|---|
0.3871 | 0.0211 | 1000 | 0.4327 | 23.4515 |
0.3696 | 0.0422 | 2000 | 0.3824 | 23.7588 |
0.3576 | 0.0632 | 3000 | 0.3596 | 22.4671 |
0.3467 | 0.0843 | 4000 | 0.3394 | 20.5368 |
0.36 | 0.1054 | 5000 | 0.3254 | 18.4529 |
0.3182 | 0.1265 | 6000 | 0.3107 | 18.0351 |
0.3293 | 0.1476 | 7000 | 0.2972 | 17.0268 |
0.2692 | 0.1686 | 8000 | 0.2888 | 18.8514 |
0.3061 | 0.1897 | 9000 | 0.2854 | 17.1612 |
0.2654 | 0.2108 | 10000 | 0.2766 | 16.8539 |
0.2954 | 0.2319 | 11000 | 0.2692 | 16.1481 |
0.2703 | 0.2529 | 12000 | 0.2632 | 15.1733 |
0.2523 | 0.2740 | 13000 | 0.2594 | 15.1541 |
0.265 | 0.2951 | 14000 | 0.2556 | 14.9669 |
0.246 | 0.3162 | 15000 | 0.2515 | 15.7784 |
0.2249 | 0.3373 | 16000 | 0.2484 | 14.2514 |
0.2478 | 0.3583 | 17000 | 0.2414 | 14.8996 |
0.2246 | 0.3794 | 18000 | 0.2432 | 14.8468 |
0.2291 | 0.4005 | 19000 | 0.2333 | 15.4038 |
0.2336 | 0.4216 | 20000 | 0.2321 | 13.8577 |
0.2448 | 0.4427 | 21000 | 0.2287 | 12.9694 |
0.2174 | 0.4637 | 22000 | 0.2237 | 13.2479 |
0.2239 | 0.4848 | 23000 | 0.2255 | 12.7197 |
0.2065 | 0.5059 | 24000 | 0.2245 | 12.4316 |
0.2467 | 0.5270 | 25000 | 0.2215 | 13.4015 |
0.2257 | 0.5480 | 26000 | 0.2200 | 14.3955 |
0.229 | 0.5691 | 27000 | 0.2172 | 12.9021 |
0.2431 | 0.5902 | 28000 | 0.2134 | 13.4927 |
0.2258 | 0.6113 | 29000 | 0.2128 | 12.1819 |
0.2025 | 0.6324 | 30000 | 0.2117 | 11.9850 |
0.2014 | 0.6534 | 31000 | 0.2055 | 11.5193 |
0.2341 | 0.6745 | 32000 | 0.2038 | 11.5097 |
0.2136 | 0.6956 | 33000 | 0.2040 | 11.4760 |
0.2348 | 0.7167 | 34000 | 0.2034 | 11.5337 |
0.1944 | 0.7378 | 35000 | 0.2005 | 11.9610 |
0.1646 | 0.7588 | 36000 | 0.2005 | 11.3656 |
0.197 | 0.7799 | 37000 | 0.1938 | 11.0199 |
0.1877 | 0.8010 | 38000 | 0.1945 | 11.7209 |
0.2122 | 0.8221 | 39000 | 0.1910 | 10.8470 |
0.1736 | 0.8432 | 40000 | 0.1905 | 11.1831 |
0.1548 | 0.8642 | 41000 | 0.1907 | 11.2215 |
0.1754 | 0.8853 | 42000 | 0.1875 | 11.1207 |
0.2125 | 0.9064 | 43000 | 0.1864 | 10.5925 |
0.1975 | 0.9275 | 44000 | 0.1834 | 11.1639 |
0.18 | 0.9485 | 45000 | 0.1827 | 10.7414 |
0.2067 | 0.9696 | 46000 | 0.1828 | 10.5589 |
0.1728 | 0.9907 | 47000 | 0.1810 | 10.5493 |
0.1253 | 1.0118 | 48000 | 0.1822 | 10.6453 |
0.1284 | 1.0329 | 49000 | 0.1818 | 10.9238 |
0.1412 | 1.0539 | 50000 | 0.1812 | 10.4677 |
0.1266 | 1.0750 | 51000 | 0.1809 | 10.6838 |
0.1093 | 1.0961 | 52000 | 0.1808 | 10.2708 |
0.1293 | 1.1172 | 53000 | 0.1809 | 10.5973 |
0.1377 | 1.1383 | 54000 | 0.1779 | 9.9443 |
0.1135 | 1.1593 | 55000 | 0.1750 | 10.0595 |
0.1029 | 1.1804 | 56000 | 0.1739 | 9.6658 |
0.0959 | 1.2015 | 57000 | 0.1776 | 10.7030 |
0.1335 | 1.2226 | 58000 | 0.1736 | 10.1268 |
0.1166 | 1.2437 | 59000 | 0.1755 | 11.6873 |
0.1079 | 1.2647 | 60000 | 0.1741 | 9.8579 |
0.124 | 1.2858 | 61000 | 0.1719 | 9.6706 |
0.1279 | 1.3069 | 62000 | 0.1725 | 10.9094 |
0.1546 | 1.3280 | 63000 | 0.1740 | 10.0643 |
0.0961 | 1.3490 | 64000 | 0.1726 | 9.6514 |
0.1167 | 1.3701 | 65000 | 0.1715 | 9.9971 |
0.1072 | 1.3912 | 66000 | 0.1691 | 9.5410 |
0.1052 | 1.4123 | 67000 | 0.1708 | 10.0067 |
0.1234 | 1.4334 | 68000 | 0.1682 | 10.5589 |
0.1131 | 1.4544 | 69000 | 0.1665 | 10.6838 |
0.1188 | 1.4755 | 70000 | 0.1668 | 11.1159 |
0.1106 | 1.4966 | 71000 | 0.1666 | 9.3777 |
0.0984 | 1.5177 | 72000 | 0.1645 | 9.6754 |
0.1206 | 1.5388 | 73000 | 0.1636 | 10.6982 |
0.1369 | 1.5598 | 74000 | 0.1625 | 9.5554 |
0.1164 | 1.5809 | 75000 | 0.1619 | 9.4881 |
0.1326 | 1.6020 | 76000 | 0.1603 | 11.4184 |
0.1168 | 1.6231 | 77000 | 0.1607 | 10.6550 |
0.107 | 1.6441 | 78000 | 0.1613 | 10.0980 |
0.1222 | 1.6652 | 79000 | 0.1611 | 9.6562 |
0.1304 | 1.6863 | 80000 | 0.1615 | 9.7186 |
0.1288 | 1.7074 | 81000 | 0.1615 | 9.6466 |
0.1101 | 1.7285 | 82000 | 0.1587 | 12.8925 |
0.1256 | 1.7495 | 83000 | 0.1577 | 11.4760 |
0.1167 | 1.7706 | 84000 | 0.1564 | 9.0992 |
0.114 | 1.7917 | 85000 | 0.1533 | 11.4424 |
0.0918 | 1.8128 | 86000 | 0.1528 | 15.9896 |
0.1338 | 1.8339 | 87000 | 0.1521 | 15.8168 |
0.1009 | 1.8549 | 88000 | 0.1514 | 15.3078 |
0.1124 | 1.8760 | 89000 | 0.1511 | 14.3330 |
0.1161 | 1.8971 | 90000 | 0.1507 | 10.4389 |
0.102 | 1.9182 | 91000 | 0.1495 | 9.4209 |
0.0921 | 1.9393 | 92000 | 0.1473 | 12.5660 |
0.1142 | 1.9603 | 93000 | 0.1477 | 11.5865 |
0.0971 | 1.9814 | 94000 | 0.1482 | 17.4493 |
0.0562 | 2.0025 | 95000 | 0.1484 | 23.0769 |
0.0697 | 2.0236 | 96000 | 0.1491 | 20.2007 |
0.0691 | 2.0446 | 97000 | 0.1487 | 14.3234 |
0.0707 | 2.0657 | 98000 | 0.1486 | 15.2694 |
0.0529 | 2.0868 | 99000 | 0.1486 | 16.5322 |
0.061 | 2.1079 | 100000 | 0.1466 | 17.9343 |
0.077 | 2.1290 | 101000 | 0.1465 | 17.1852 |
0.0748 | 2.1500 | 102000 | 0.1474 | 15.5767 |
0.0624 | 2.1711 | 103000 | 0.1471 | 15.2118 |
0.0625 | 2.1922 | 104000 | 0.1452 | 27.2352 |
0.0876 | 2.2133 | 105000 | 0.1476 | 27.5137 |
0.0683 | 2.2344 | 106000 | 0.1468 | 20.1911 |
0.0539 | 2.2554 | 107000 | 0.1459 | 19.6101 |
0.0627 | 2.2765 | 108000 | 0.1462 | 19.4997 |
0.0548 | 2.2976 | 109000 | 0.1469 | 18.6546 |
0.0559 | 2.3187 | 110000 | 0.1453 | 15.9224 |
0.0667 | 2.3397 | 111000 | 0.1447 | 20.4312 |
0.0611 | 2.3608 | 112000 | 0.1442 | 19.0963 |
0.0672 | 2.3819 | 113000 | 0.1441 | 19.7349 |
0.0517 | 2.4030 | 114000 | 0.1435 | 17.6894 |
0.0584 | 2.4241 | 115000 | 0.1439 | 21.5884 |
0.0634 | 2.4451 | 116000 | 0.1428 | 22.2942 |
0.0754 | 2.4662 | 117000 | 0.1420 | 25.4346 |
0.0537 | 2.4873 | 118000 | 0.1413 | 29.4440 |
0.0478 | 2.5084 | 119000 | 0.1412 | 21.6796 |
0.0509 | 2.5295 | 120000 | 0.1414 | 22.2414 |
0.0749 | 2.5505 | 121000 | 0.1405 | 18.5153 |
0.069 | 2.5716 | 122000 | 0.1391 | 17.9679 |
0.0614 | 2.5927 | 123000 | 0.1395 | 19.7157 |
0.0628 | 2.6138 | 124000 | 0.1382 | 19.7926 |
0.0518 | 2.6349 | 125000 | 0.1390 | 19.3172 |
0.078 | 2.6559 | 126000 | 0.1379 | 23.3458 |
0.0578 | 2.6770 | 127000 | 0.1388 | 21.2427 |
0.0406 | 2.6981 | 128000 | 0.1384 | 24.9832 |
0.0494 | 2.7192 | 129000 | 0.1373 | 21.6124 |
0.0714 | 2.7402 | 130000 | 0.1375 | 22.4671 |
0.0646 | 2.7613 | 131000 | 0.1369 | 23.7203 |
0.0582 | 2.7824 | 132000 | 0.1372 | 22.2462 |
0.0594 | 2.8035 | 133000 | 0.1368 | 22.1646 |
0.0435 | 2.8246 | 134000 | 0.1364 | 21.4684 |
0.0509 | 2.8456 | 135000 | 0.1361 | 19.0243 |
0.0553 | 2.8667 | 136000 | 0.1365 | 21.0506 |
0.0716 | 2.8878 | 137000 | 0.1360 | 21.2859 |
0.0621 | 2.9089 | 138000 | 0.1359 | 19.7926 |
0.0597 | 2.9300 | 139000 | 0.1358 | 21.4011 |
0.0476 | 2.9510 | 140000 | 0.1357 | 21.3195 |
0.0483 | 2.9721 | 141000 | 0.1355 | 21.3195 |
0.0504 | 2.9932 | 142000 | 0.1355 | 22.1166 |
0.1197 | 2.8636 | 143000 | 0.1315 | 21.4829 |
0.1389 | 2.8836 | 144000 | 0.1266 | 22.1632 |
0.1242 | 2.9036 | 145000 | 0.1244 | 23.1137 |
0.1355 | 2.9236 | 146000 | 0.1228 | 21.0505 |
0.1257 | 2.9437 | 147000 | 0.1218 | 20.4334 |
0.1027 | 2.9637 | 148000 | 0.1212 | 21.6541 |
0.1186 | 2.9837 | 149000 | 0.1208 | 22.6497 |
Framework versions
- Transformers 4.47.0
- Pytorch 2.5.1+cu121
- Datasets 3.1.0
- Tokenizers 0.21.0
- Downloads last month
- 12
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.