Whisper tiny Portuguese English
This model is a fine-tuned version of openai/whisper-tiny on the Portuguese English dataset. It achieves the following results on the evaluation set:
- Loss: 0.3207
- Wer: 13.4949
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: 3.75e-05
- train_batch_size: 32
- eval_batch_size: 64
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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: 50
- training_steps: 1000
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.2110 | 1.4706 | 100 | 0.2759 | 14.1473 |
| 0.0993 | 2.9412 | 200 | 0.2859 | 13.1780 |
| 0.0457 | 4.4118 | 300 | 0.2974 | 12.5815 |
| 0.0288 | 5.8824 | 400 | 0.3041 | 13.0848 |
| 0.0207 | 7.3529 | 500 | 0.3109 | 13.3271 |
| 0.0177 | 8.8235 | 600 | 0.3217 | 13.0848 |
| 0.0139 | 10.2941 | 700 | 0.3202 | 13.4390 |
| 0.0142 | 11.7647 | 800 | 0.3234 | 13.4203 |
| 0.0144 | 13.2353 | 900 | 0.3200 | 13.2526 |
| 0.0114 | 14.7059 | 1000 | 0.3207 | 13.4949 |
Framework versions
- Transformers 5.23.0.dev0
- Pytorch 2.11.0+cu130
- Datasets 5.1.0
- Tokenizers 0.23.3