Whisper large v2 aphasia - 8000
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: 0.5093
- Wer Ortho: 24.2302
- Wer: 23.3912
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-06
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: constant_with_warmup
- training_steps: 2000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer Ortho | Wer |
|---|---|---|---|---|---|
| 0.1923 | 0.0943 | 400 | 0.5208 | 24.4866 | 23.7810 |
| 0.2205 | 0.1886 | 800 | 0.5260 | 24.6175 | 24.0092 |
| 0.1873 | 0.2830 | 1200 | 0.5211 | 24.1294 | 23.5221 |
| 0.2128 | 0.3773 | 1600 | 0.5311 | 24.0080 | 23.2862 |
| 0.2535 | 0.4716 | 2000 | 0.5093 | 24.2302 | 23.3912 |
Framework versions
- Transformers 4.57.3
- Pytorch 2.9.0+cu126
- Datasets 3.6.0
- Tokenizers 0.22.1