Whisper Small Ka - tsobolev
This model is a fine-tuned version of openai/whisper-small on the CV14 dataset. It achieves the following results on the test set:
- Loss: 0.0542
- Wer Ortho: 0.2785
- Wer: 0.2470
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
Training - Mozilla Common Voice 14 train split Validation - Mozilla Common Voice 14 test split
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: 50
- training_steps: 5000
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer Ortho | Wer |
|---|---|---|---|---|---|
| 0.0123 | 2.76 | 5000 | 0.0542 | 0.2785 | 0.2470 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.0.0
- Datasets 2.14.4
- Tokenizers 0.13.3