Whisper Small Co 5
This model is a fine-tuned version of openai/whisper-small on the Co audio dataset dataset. It achieves the following results on the evaluation set:
- Loss: 0.0017
- Cer Ortho: 0.0
- Cer: 0.0
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: 16
- eval_batch_size: 16
- 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: constant_with_warmup
- lr_scheduler_warmup_steps: 50
- training_steps: 1000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Cer Ortho | Cer |
|---|---|---|---|---|---|
| 1.8472 | 1.2195 | 100 | 0.8674 | 40.2730 | 38.8699 |
| 0.4906 | 2.4390 | 200 | 0.3698 | 22.8669 | 21.9178 |
| 0.2826 | 3.6585 | 300 | 0.1373 | 9.3857 | 7.8767 |
| 0.1323 | 4.8780 | 400 | 0.0503 | 3.5836 | 3.4247 |
| 0.0573 | 6.0976 | 500 | 0.0279 | 1.3652 | 1.1986 |
| 0.0241 | 7.3171 | 600 | 0.0413 | 1.7065 | 1.3699 |
| 0.0214 | 8.5366 | 700 | 0.0129 | 3.5836 | 3.4247 |
| 0.0116 | 9.7561 | 800 | 0.0044 | 0.0 | 0.0 |
| 0.0139 | 10.9756 | 900 | 0.0061 | 0.1706 | 0.1712 |
| 0.0161 | 12.1951 | 1000 | 0.0017 | 0.0 | 0.0 |
Framework versions
- Transformers 4.50.3
- Pytorch 2.6.0+cu124
- Datasets 3.5.0
- Tokenizers 0.21.1