wh_small_cv_common_voices_52393__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: 0.2983
- Global Wer: 40.3623
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 |
|---|---|---|---|---|
| 1.0107 | 0.0411 | 50 | 0.8289 | 57.5260 |
| 0.6657 | 0.0821 | 100 | 0.6580 | 50.5943 |
| 0.5927 | 0.1232 | 150 | 0.5769 | 46.9064 |
| 0.4906 | 0.1642 | 200 | 0.5067 | 44.9602 |
| 0.4491 | 0.2053 | 250 | 0.4597 | 44.0066 |
| 0.3647 | 0.2463 | 300 | 0.4000 | 42.4435 |
| 0.3391 | 0.2874 | 350 | 0.3424 | 40.8978 |
| 0.2915 | 0.3284 | 400 | 0.3106 | 40.0967 |
| 0.2943 | 0.3695 | 450 | 0.3004 | 39.6351 |
| 0.2841 | 0.4105 | 500 | 0.2983 | 40.3623 |
Framework versions
- Transformers 5.0.0.dev0
- Pytorch 2.9.0+cu126
- Datasets 3.6.0
- Tokenizers 0.22.2