Whisper Medium v3
This model is a fine-tuned version of openai/whisper-medium on the Common Voice 13 dataset. It achieves the following results on the evaluation set:
- Loss: 0.1501
- Wer Ortho: 0.1759
- Wer: 0.0791
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: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 50
- training_steps: 4000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer Ortho | Wer |
|---|---|---|---|---|---|
| 0.2054 | 0.03 | 500 | 0.2392 | 0.2715 | 0.1484 |
| 0.1782 | 0.05 | 1000 | 0.2056 | 0.2411 | 0.1155 |
| 0.1736 | 0.08 | 1500 | 0.1768 | 0.1990 | 0.0994 |
| 0.1662 | 0.11 | 2000 | 0.1677 | 0.1925 | 0.0940 |
| 0.1409 | 0.13 | 2500 | 0.1589 | 0.1891 | 0.0860 |
| 0.1346 | 0.16 | 3000 | 0.1565 | 0.1897 | 0.0881 |
| 0.1263 | 0.19 | 3500 | 0.1523 | 0.1805 | 0.0819 |
| 0.137 | 0.22 | 4000 | 0.1501 | 0.1759 | 0.0791 |
Framework versions
- Transformers 4.37.2
- Pytorch 2.0.1+cu117
- Datasets 2.16.1
- Tokenizers 0.15.1