Whisper Large V2
This model is a fine-tuned version of openai/whisper-large-v2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.3157
- Wer: 10.7746
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: 3e-05
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 20
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.5562 | 0.49 | 30 | 0.3054 | 11.4500 |
| 0.2771 | 0.98 | 60 | 0.2629 | 11.3656 |
| 0.1427 | 1.48 | 90 | 0.2726 | 13.1173 |
| 0.1369 | 1.97 | 120 | 0.2639 | 10.9751 |
| 0.0638 | 2.46 | 150 | 0.2741 | 11.9038 |
| 0.0541 | 2.95 | 180 | 0.2833 | 10.0992 |
| 0.0289 | 3.44 | 210 | 0.3024 | 10.7851 |
| 0.0198 | 3.93 | 240 | 0.3073 | 10.6902 |
| 0.0103 | 4.43 | 270 | 0.3177 | 10.4158 |
| 0.0089 | 4.92 | 300 | 0.3157 | 10.7746 |
Framework versions
- Transformers 4.38.0.dev0
- Pytorch 2.1.0+cu121
- Datasets 2.14.6
- Tokenizers 0.15.0
- Downloads last month
- 3