whisper-tiny-kor_eng_tiny_ed_pr
This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2603
- Cer: 3.4043
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: 3e-05
- train_batch_size: 12
- eval_batch_size: 6
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- training_steps: 4000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Cer |
|---|---|---|---|---|
| 0.2505 | 25.0 | 100 | 0.2657 | 6.3830 |
| 0.0006 | 50.0 | 200 | 0.2550 | 6.8085 |
| 0.0001 | 75.0 | 300 | 0.2495 | 3.4043 |
| 0.0001 | 100.0 | 400 | 0.2514 | 3.4043 |
| 0.0 | 125.0 | 500 | 0.2495 | 3.4043 |
| 0.0 | 150.0 | 600 | 0.2493 | 3.4043 |
| 0.0 | 175.0 | 700 | 0.2504 | 3.4043 |
| 0.0 | 200.0 | 800 | 0.2511 | 3.4043 |
| 0.0 | 225.0 | 900 | 0.2522 | 3.4043 |
| 0.0 | 250.0 | 1000 | 0.2517 | 3.4043 |
| 0.0 | 275.0 | 1100 | 0.2516 | 3.4043 |
| 0.0 | 300.0 | 1200 | 0.2527 | 3.4043 |
| 0.0 | 325.0 | 1300 | 0.2533 | 3.4043 |
| 0.0 | 350.0 | 1400 | 0.2541 | 3.4043 |
| 0.0 | 375.0 | 1500 | 0.2550 | 3.4043 |
| 0.0 | 400.0 | 1600 | 0.2549 | 3.4043 |
| 0.0 | 425.0 | 1700 | 0.2542 | 3.4043 |
| 0.0 | 450.0 | 1800 | 0.2546 | 3.4043 |
| 0.0 | 475.0 | 1900 | 0.2541 | 3.4043 |
| 0.0 | 500.0 | 2000 | 0.2547 | 3.4043 |
| 0.0 | 525.0 | 2100 | 0.2563 | 3.4043 |
| 0.0 | 550.0 | 2200 | 0.2562 | 3.4043 |
| 0.0 | 575.0 | 2300 | 0.2573 | 3.4043 |
| 0.0 | 600.0 | 2400 | 0.2577 | 3.4043 |
| 0.0 | 625.0 | 2500 | 0.2575 | 3.4043 |
| 0.0 | 650.0 | 2600 | 0.2591 | 3.4043 |
| 0.0 | 675.0 | 2700 | 0.2586 | 3.4043 |
| 0.0 | 700.0 | 2800 | 0.2590 | 3.4043 |
| 0.0 | 725.0 | 2900 | 0.2597 | 3.4043 |
| 0.0 | 750.0 | 3000 | 0.2594 | 3.4043 |
| 0.0 | 775.0 | 3100 | 0.2592 | 3.4043 |
| 0.0 | 800.0 | 3200 | 0.2598 | 3.4043 |
| 0.0 | 825.0 | 3300 | 0.2599 | 3.4043 |
| 0.0 | 850.0 | 3400 | 0.2608 | 3.4043 |
| 0.0 | 875.0 | 3500 | 0.2598 | 3.4043 |
| 0.0 | 900.0 | 3600 | 0.2603 | 3.4043 |
| 0.0 | 925.0 | 3700 | 0.2595 | 3.4043 |
| 0.0 | 950.0 | 3800 | 0.2597 | 3.4043 |
| 0.0 | 975.0 | 3900 | 0.2599 | 3.4043 |
| 0.0 | 1000.0 | 4000 | 0.2603 | 3.4043 |
Framework versions
- Transformers 4.51.3
- Pytorch 2.6.0+cu126
- Datasets 3.5.0
- Tokenizers 0.21.1
- Downloads last month
- 3