whisper-small-common-voice-2
This model is a fine-tuned version of lejonck/whisper-small-common-voice-1 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0936
- Wer: 0.2679
- Cer: 0.3627
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: 12
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|---|---|---|---|---|---|
| 0.3127 | 1.0 | 750 | 0.0936 | 0.2679 | 0.3627 |
| 0.1023 | 2.0 | 1500 | 0.0309 | 0.3188 | 0.3846 |
| 0.0247 | 3.0 | 2250 | 0.0098 | 0.3269 | 0.3831 |
| 0.0133 | 4.0 | 3000 | 0.0048 | 0.3528 | 0.3772 |
| 0.0084 | 5.0 | 3750 | 0.0027 | 0.2966 | 0.3646 |
| 0.0021 | 6.0 | 4500 | 0.0016 | 0.4000 | 0.3859 |
Framework versions
- Transformers 4.55.2
- Pytorch 2.7.0+cu126
- Datasets 2.19.1
- Tokenizers 0.21.4