xlsr53-mupe-3
This model is a fine-tuned version of lejonck/xlsr53-mupe-2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.6326
- Wer: 0.2721
- Cer: 0.1210
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 |
|---|---|---|---|---|---|
| 0.6489 | 1.0 | 4000 | 0.6075 | 0.2772 | 0.1215 |
| 0.5594 | 2.0 | 8000 | 0.6086 | 0.2740 | 0.1213 |
| 0.9263 | 3.0 | 12000 | 0.6326 | 0.2721 | 0.1209 |
| 0.5107 | 4.0 | 16000 | 0.6452 | 0.2758 | 0.1216 |
| 0.1879 | 5.0 | 20000 | 0.6454 | 0.2758 | 0.1221 |
| 0.5059 | 6.0 | 24000 | 0.6452 | 0.2764 | 0.1218 |
Framework versions
- Transformers 4.55.2
- Pytorch 2.7.0+cu126
- Datasets 4.0.0
- Tokenizers 0.21.4