wh_small_mucs_30_70_mucs_52508__val_finetune_trial
This model is a fine-tuned version of openai/whisper-small on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.4102
- Global Wer: 40.1558
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: 5e-06
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Use 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
- lr_scheduler_warmup_steps: 50
- training_steps: 500
Training results
| Training Loss | Epoch | Step | Validation Loss | Global Wer |
|---|---|---|---|---|
| 1.5540 | 0.0133 | 50 | 1.0859 | 55.1433 |
| 1.0086 | 0.0266 | 100 | 0.8461 | 52.3995 |
| 0.9141 | 0.0399 | 150 | 0.7470 | 45.6799 |
| 0.7475 | 0.0531 | 200 | 0.6591 | 60.1264 |
| 0.5349 | 0.0664 | 250 | 0.5893 | 44.6889 |
| 0.5338 | 0.0797 | 300 | 0.5298 | 44.7005 |
| 0.4313 | 0.0930 | 350 | 0.4720 | 44.8404 |
| 0.4111 | 0.1063 | 400 | 0.4396 | 45.6978 |
| 0.3785 | 0.1196 | 450 | 0.4267 | 42.0547 |
| 0.3950 | 0.1329 | 500 | 0.4102 | 40.1558 |
Framework versions
- Transformers 5.0.0.dev0
- Pytorch 2.9.0+cu126
- Datasets 3.6.0
- Tokenizers 0.22.2