wh_small_mucs_30_70_no_lg_id_mucs_52580__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: 0.3483
- Global Wer: 30.7541
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 | 1.1463 | 50.2135 |
| 1.0086 | 0.0266 | 100 | 0.8641 | 47.8086 |
| 0.9142 | 0.0399 | 150 | 0.7299 | 37.2653 |
| 0.7471 | 0.0531 | 200 | 0.6395 | 48.3989 |
| 0.5347 | 0.0664 | 250 | 0.5561 | 36.0846 |
| 0.5334 | 0.0797 | 300 | 0.4782 | 35.4050 |
| 0.4312 | 0.0930 | 350 | 0.4059 | 35.2525 |
| 0.4112 | 0.1063 | 400 | 0.3810 | 35.9212 |
| 0.3778 | 0.1196 | 450 | 0.3628 | 31.7562 |
| 0.3950 | 0.1329 | 500 | 0.3483 | 30.7541 |
Framework versions
- Transformers 5.0.0.dev0
- Pytorch 2.9.0+cu126
- Datasets 3.6.0
- Tokenizers 0.22.2