xlsr53-mupe-2
This model is a fine-tuned version of lejonck/xlsr53-mupe-1 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.7844
- Wer: 0.3075
- Cer: 0.1462
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: 2
- eval_batch_size: 2
- seed: 42
- optimizer: Use OptimizerNames.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: 100
- num_epochs: 12
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|---|---|---|---|---|---|
| 0.557 | 1.0 | 3000 | 0.7895 | 0.3494 | 0.1662 |
| 0.5848 | 2.0 | 6000 | 0.7954 | 0.3314 | 0.1557 |
| 1.278 | 3.0 | 9000 | 0.7741 | 0.3238 | 0.1526 |
| 0.9106 | 4.0 | 12000 | 0.7581 | 0.3246 | 0.1523 |
| 0.7749 | 5.0 | 15000 | 0.7562 | 0.3164 | 0.1501 |
| 0.3564 | 6.0 | 18000 | 0.7690 | 0.3139 | 0.1475 |
| 0.5751 | 7.0 | 21000 | 0.7782 | 0.3166 | 0.1496 |
| 0.465 | 8.0 | 24000 | 0.7731 | 0.3117 | 0.1480 |
| 0.3877 | 9.0 | 27000 | 0.7701 | 0.3095 | 0.1465 |
| 0.1417 | 10.0 | 30000 | 0.7805 | 0.3096 | 0.1470 |
| 0.4942 | 11.0 | 33000 | 0.7854 | 0.3093 | 0.1467 |
| 0.2666 | 12.0 | 36000 | 0.7844 | 0.3078 | 0.1462 |
Framework versions
- Transformers 4.55.2
- Pytorch 2.7.0+cu126
- Datasets 4.0.0
- Tokenizers 0.21.4