wh_small_cv_no_lg_id_common_voices_52395__val_finetune_trial
This model is a fine-tuned version of openai/whisper-small on the common_voice_22_0 dataset. It achieves the following results on the evaluation set:
- Loss: 0.3017
- Global Wer: 45.2649
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 |
|---|---|---|---|---|
| 0.7614 | 0.0411 | 50 | 0.5752 | 61.8278 |
| 0.4647 | 0.0821 | 100 | 0.4477 | 54.0123 |
| 0.4257 | 0.1232 | 150 | 0.4020 | 49.9891 |
| 0.3572 | 0.1642 | 200 | 0.3679 | 47.2373 |
| 0.3408 | 0.2053 | 250 | 0.3579 | 47.3636 |
| 0.2975 | 0.2463 | 300 | 0.3368 | 46.2011 |
| 0.3103 | 0.2874 | 350 | 0.3212 | 44.8818 |
| 0.2941 | 0.3284 | 400 | 0.3106 | 44.9863 |
| 0.2977 | 0.3695 | 450 | 0.3041 | 43.8107 |
| 0.2895 | 0.4105 | 500 | 0.3017 | 45.2649 |
Framework versions
- Transformers 5.0.0.dev0
- Pytorch 2.9.0+cu126
- Datasets 3.6.0
- Tokenizers 0.22.2