wav2vec2-large-xlsr-grosman-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 |
|---|
| 0.3647 | 1.0 | 101 | 0.4522 | 0.1019 |
| 0.2988 | 2.0 | 203 | 0.4142 | 0.1109 |
| 0.2854 | 3.0 | 304 | 0.4568 | 0.1043 |
| 0.2617 | 4.0 | 406 | 0.4920 | 0.1016 |
| 0.2443 | 5.0 | 507 | 0.4380 | 0.0989 |
| 0.259 | 6.0 | 609 | 0.4550 | 0.0952 |
| 0.2425 | 7.0 | 710 | 0.4414 | 0.0891 |
| 0.2294 | 8.0 | 812 | 0.4546 | 0.0902 |
| 0.2241 | 9.0 | 913 | 0.4846 | 0.0904 |
| 0.2218 | 10.0 | 1015 | 0.4858 | 0.0866 |
Framework versions
- Transformers 4.28.0
- Pytorch 2.6.0+cu124
- Datasets 4.0.0
- Tokenizers 0.13.3
