whisper-small-ptbr6
This model is a fine-tuned version of openai/whisper-small on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.7704
- Wer: 0.3545
- Cer: 0.5847
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: 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: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|---|---|---|---|---|---|
| 0.3936 | 1.0 | 250 | 0.7103 | 0.3557 | 0.5853 |
| 0.1695 | 2.0 | 500 | 0.7217 | 0.3633 | 0.5874 |
| 0.0699 | 3.0 | 750 | 0.7670 | 0.3539 | 0.5858 |
| 0.0231 | 4.0 | 1000 | 0.8688 | 0.3685 | 0.5935 |
| 0.0051 | 5.0 | 1250 | 0.8842 | 0.3656 | 0.5904 |
| 0.0015 | 6.0 | 1500 | 0.8804 | 0.3551 | 0.5874 |
| 0.003 | 7.0 | 1750 | 0.8941 | 0.3609 | 0.5876 |
| 0.0006 | 8.0 | 2000 | 0.8938 | 0.3609 | 0.5877 |
| 0.0006 | 9.0 | 2250 | 0.9445 | 0.3586 | 0.5870 |
Framework versions
- Transformers 4.55.0
- Pytorch 2.7.0+cu126
- Datasets 4.0.0
- Tokenizers 0.21.4