whisper-small-sw-common-voice-ndizi-158-50epochs
This model is a fine-tuned version of smutuvi/whisper-small-sw-common-voice on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.6112
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: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 50
- num_epochs: 50
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| No log | 1.0 | 18 | 1.9359 |
| 1.7712 | 2.0 | 36 | 1.9211 |
| 1.7254 | 3.0 | 54 | 1.8970 |
| 1.7254 | 4.0 | 72 | 1.8737 |
| 1.6754 | 5.0 | 90 | 1.8498 |
| 1.6138 | 6.0 | 108 | 1.8281 |
| 1.5935 | 7.0 | 126 | 1.8054 |
| 1.5935 | 8.0 | 144 | 1.7885 |
| 1.5657 | 9.0 | 162 | 1.7707 |
| 1.4709 | 10.0 | 180 | 1.7551 |
| 1.4709 | 11.0 | 198 | 1.7389 |
| 1.5099 | 12.0 | 216 | 1.7300 |
| 1.5411 | 13.0 | 234 | 1.7178 |
| 1.4451 | 14.0 | 252 | 1.7097 |
| 1.4451 | 15.0 | 270 | 1.7003 |
| 1.3941 | 16.0 | 288 | 1.6966 |
| 1.403 | 17.0 | 306 | 1.6892 |
| 1.403 | 18.0 | 324 | 1.6833 |
| 1.434 | 19.0 | 342 | 1.6766 |
| 1.377 | 20.0 | 360 | 1.6718 |
| 1.349 | 21.0 | 378 | 1.6680 |
| 1.349 | 22.0 | 396 | 1.6610 |
| 1.3351 | 23.0 | 414 | 1.6592 |
| 1.421 | 24.0 | 432 | 1.6527 |
| 1.3146 | 25.0 | 450 | 1.6508 |
| 1.3146 | 26.0 | 468 | 1.6470 |
| 1.3393 | 27.0 | 486 | 1.6451 |
| 1.3039 | 28.0 | 504 | 1.6394 |
| 1.3039 | 29.0 | 522 | 1.6391 |
| 1.3886 | 30.0 | 540 | 1.6354 |
| 1.2247 | 31.0 | 558 | 1.6345 |
| 1.2959 | 32.0 | 576 | 1.6306 |
| 1.2959 | 33.0 | 594 | 1.6292 |
| 1.3447 | 34.0 | 612 | 1.6266 |
| 1.2708 | 35.0 | 630 | 1.6246 |
| 1.2708 | 36.0 | 648 | 1.6217 |
| 1.2882 | 37.0 | 666 | 1.6229 |
| 1.2963 | 38.0 | 684 | 1.6186 |
| 1.2696 | 39.0 | 702 | 1.6202 |
| 1.2696 | 40.0 | 720 | 1.6196 |
| 1.2019 | 41.0 | 738 | 1.6164 |
| 1.3452 | 42.0 | 756 | 1.6157 |
| 1.3452 | 43.0 | 774 | 1.6145 |
| 1.2787 | 44.0 | 792 | 1.6154 |
| 1.2513 | 45.0 | 810 | 1.6120 |
| 1.2227 | 46.0 | 828 | 1.6146 |
| 1.2227 | 47.0 | 846 | 1.6120 |
| 1.2523 | 48.0 | 864 | 1.6125 |
| 1.2498 | 49.0 | 882 | 1.6119 |
| 1.3094 | 50.0 | 900 | 1.6112 |
Framework versions
- PEFT 0.7.2.dev0
- Transformers 4.37.0.dev0
- Pytorch 2.0.0
- Datasets 2.16.1
- Tokenizers 0.15.0