Whisper Small MN with custom data - Zagi
This model is a fine-tuned version of openai/whisper-tiny on the audiofolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.0917
- Wer: 9.3784
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: 1e-05
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- training_steps: 4000
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.0653 | 0.61 | 500 | 0.1102 | 13.5820 |
| 0.054 | 1.21 | 1000 | 0.1002 | 11.9380 |
| 0.0523 | 1.82 | 1500 | 0.0966 | 11.5903 |
| 0.0366 | 2.43 | 2000 | 0.0954 | 10.9710 |
| 0.0168 | 3.03 | 2500 | 0.0909 | 10.3866 |
| 0.0204 | 3.64 | 3000 | 0.0912 | 9.7817 |
| 0.0067 | 4.25 | 3500 | 0.0910 | 9.4936 |
| 0.0078 | 4.85 | 4000 | 0.0917 | 9.3784 |
Framework versions
- Transformers 4.39.0
- Pytorch 2.0.1+cu117
- Datasets 2.18.0
- Tokenizers 0.15.2