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: 0.4367
- Wer: 13.2014
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 |
|---|---|---|---|---|
| 0.7481 | 0.55 | 30 | 0.4470 | 24.0337 |
| 0.3791 | 1.09 | 60 | 0.3935 | 17.3940 |
| 0.2077 | 1.64 | 90 | 0.3841 | 14.4015 |
| 0.1739 | 2.18 | 120 | 0.3804 | 14.5729 |
| 0.0918 | 2.73 | 150 | 0.4027 | 15.1808 |
| 0.0684 | 3.27 | 180 | 0.4156 | 15.3367 |
| 0.0391 | 3.82 | 210 | 0.4038 | 15.5393 |
| 0.0197 | 4.36 | 240 | 0.4326 | 13.5287 |
| 0.0128 | 4.91 | 270 | 0.4367 | 13.2014 |
Framework versions
- Transformers 4.38.0.dev0
- Pytorch 2.1.0+cu121
- Datasets 2.14.6
- Tokenizers 0.15.0
- Downloads last month
- 1