whisper-small-uk-v2
This model is a fine-tuned version of nikes64/whisper-small-uk on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.1369
- Wer: 11.2911
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
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- 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 | Wer |
|---|---|---|---|---|
| 0.087 | 1.4 | 1000 | 0.1092 | 11.8439 |
| 0.0429 | 2.8 | 2000 | 0.1035 | 11.0597 |
| 0.0132 | 4.2 | 3000 | 0.1247 | 11.1431 |
| 0.0067 | 5.6 | 4000 | 0.1369 | 11.2911 |
Framework versions
- Transformers 4.38.0.dev0
- Pytorch 2.2.0+cu121
- Datasets 2.17.1
- Tokenizers 0.15.2