wav2vec2-large-xlsr-facebook-53-aug-words-phoneme-exp-1-v5
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
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: 3e-05
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 20
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Per |
|---|
| 0.5793 | 1.0 | 101 | 0.4673 | 0.1784 |
| 0.4621 | 2.0 | 203 | 0.4753 | 0.1820 |
| 0.4421 | 3.0 | 304 | 0.4730 | 0.1759 |
| 0.4098 | 4.0 | 406 | 0.5114 | 0.1795 |
| 0.3772 | 5.0 | 507 | 0.4832 | 0.1823 |
| 0.3841 | 6.0 | 609 | 0.4857 | 0.1894 |
| 0.3606 | 7.0 | 710 | 0.4667 | 0.1784 |
| 0.3346 | 8.0 | 812 | 0.4614 | 0.1759 |
| 0.3186 | 9.0 | 913 | 0.5453 | 0.1818 |
| 0.3277 | 10.0 | 1015 | 0.4911 | 0.1825 |
| 0.2913 | 11.0 | 1116 | 0.4800 | 0.1843 |
| 0.2976 | 12.0 | 1218 | 0.4374 | 0.1773 |
| 0.2771 | 13.0 | 1319 | 0.4365 | 0.1782 |
| 0.2863 | 14.0 | 1421 | 0.4304 | 0.1700 |
| 0.2829 | 15.0 | 1522 | 0.4670 | 0.1716 |
| 0.2762 | 16.0 | 1624 | 0.4907 | 0.1713 |
| 0.2953 | 17.0 | 1725 | 0.4728 | 0.1688 |
| 0.2674 | 18.0 | 1827 | 0.4558 | 0.1709 |
| 0.265 | 19.0 | 1928 | 0.4576 | 0.1706 |
| 0.2733 | 19.9 | 2020 | 0.4551 | 0.1704 |
Framework versions
- Transformers 4.28.0
- Pytorch 2.6.0+cu124
- Datasets 4.0.0
- Tokenizers 0.13.3
