Whisper Base zh-TW
This model is a fine-tuned version of openai/whisper-base on the Common Voice 13.0 dataset. It achieves the following results on the evaluation set:
- Loss: 0.3403
- Cer: 16.6369
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
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Cer |
|---|---|---|---|---|
| 0.0772 | 1.38 | 1000 | 0.3230 | 17.4367 |
| 0.0436 | 2.75 | 2000 | 0.3191 | 16.4661 |
| 0.0111 | 4.13 | 3000 | 0.3343 | 16.5334 |
| 0.0078 | 5.5 | 4000 | 0.3403 | 16.6369 |
Framework versions
- Transformers 4.38.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2
- Downloads last month
- 11
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.
Model tree for JunWorks/whisper-base-zhTW
Base model
openai/whisper-base