wh_med_fleurs_fleurs_52376__val_finetune_trial
This model is a fine-tuned version of openai/whisper-medium on the fleurs dataset. It achieves the following results on the evaluation set:
- Loss: 0.2269
- Global Wer: 24.2644
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 |
|---|---|---|---|---|
| 0.5990 | 0.0945 | 50 | 0.5550 | 36.9088 |
| 0.4468 | 0.1890 | 100 | 0.3957 | 33.1099 |
| 0.2551 | 0.2836 | 150 | 0.2852 | 29.5903 |
| 0.2599 | 0.3781 | 200 | 0.2739 | 29.1248 |
| 0.2369 | 0.4726 | 250 | 0.2602 | 26.8901 |
| 0.2156 | 0.5671 | 300 | 0.2505 | 26.2197 |
| 0.2195 | 0.6616 | 350 | 0.2369 | 25.1397 |
| 0.2015 | 0.7561 | 400 | 0.2354 | 24.7486 |
| 0.1834 | 0.8507 | 450 | 0.2261 | 24.2086 |
| 0.1927 | 0.9452 | 500 | 0.2269 | 24.2644 |
Framework versions
- Transformers 5.0.0.dev0
- Pytorch 2.9.0+cu126
- Datasets 3.6.0
- Tokenizers 0.22.2