wh_small_fleurs_fleurs_52392__val_finetune_trial
This model is a fine-tuned version of openai/whisper-small on the fleurs dataset. It achieves the following results on the evaluation set:
- Loss: 0.3070
- Global Wer: 33.4264
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 |
|---|---|---|---|---|
| 0.7979 | 0.0945 | 50 | 0.6584 | 50.9683 |
| 0.5752 | 0.1890 | 100 | 0.5427 | 47.1508 |
| 0.4703 | 0.2836 | 150 | 0.4826 | 40.8939 |
| 0.4566 | 0.3781 | 200 | 0.4492 | 39.4972 |
| 0.4171 | 0.4726 | 250 | 0.4190 | 37.4488 |
| 0.3705 | 0.5671 | 300 | 0.3932 | 37.4115 |
| 0.3435 | 0.6616 | 350 | 0.3637 | 34.8417 |
| 0.3216 | 0.7561 | 400 | 0.3474 | 34.4879 |
| 0.3001 | 0.8507 | 450 | 0.3243 | 34.0410 |
| 0.2756 | 0.9452 | 500 | 0.3070 | 33.4264 |
Framework versions
- Transformers 5.0.0.dev0
- Pytorch 2.9.0+cu126
- Datasets 3.6.0
- Tokenizers 0.22.2