Whisper Small id - convonce
This model is a fine-tuned version of openai/whisper-small on the Common Voice 17.0 dataset. It achieves the following results on the evaluation set:
- Loss: 0.3414
- Wer: 17.3697
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: 8
- seed: 42
- optimizer: Use 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: 500
- training_steps: 5000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.2139 | 1.9231 | 1000 | 0.2519 | 17.5089 |
| 0.0278 | 3.8462 | 2000 | 0.2925 | 17.9359 |
| 0.0052 | 5.7692 | 3000 | 0.3140 | 17.5043 |
| 0.0017 | 7.6923 | 4000 | 0.3356 | 17.4857 |
| 0.0013 | 9.6154 | 5000 | 0.3414 | 17.3697 |
Framework versions
- Transformers 4.48.3
- Pytorch 2.5.1+cu121
- Datasets 3.3.0
- Tokenizers 0.21.0