Whisper Medium Japanese
This model is a fine-tuned version of openai/whisper-medium on the mozilla-foundation/common_voice_11_0 ja dataset. It achieves the following results on the evaluation set:
- Loss: 0.2165
- Wer: 62.6897
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: 2
- eval_batch_size: 1
- 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: 5000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.2264 | 0.2 | 1000 | 0.3102 | 79.3588 |
| 0.3195 | 0.4 | 2000 | 0.2830 | 78.1955 |
| 0.3905 | 0.6 | 3000 | 0.2508 | 72.9181 |
| 0.2478 | 0.8 | 4000 | 0.2407 | 68.8466 |
| 0.0922 | 1.1 | 5000 | 0.2165 | 62.6897 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu117
- Datasets 2.7.1.dev0
- Tokenizers 0.13.2