Whisper Base Chinese-Mandarin
This model is a fine-tuned version of xmzhu/whisper-tiny-zh on the mozilla-foundation/common_voice_16_0 zh-CN dataset. It achieves the following results on the evaluation set:
- Loss: 0.5759
- Wer: 91.1266
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: 5e-07
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 200
- training_steps: 1000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.6689 | 0.2 | 200 | 0.5854 | 91.6311 |
| 0.6314 | 1.07 | 400 | 0.5791 | 91.1788 |
| 0.653 | 1.27 | 600 | 0.5759 | 91.1266 |
| 0.699 | 2.13 | 800 | 0.5749 | 91.2049 |
| 0.5613 | 3.0 | 1000 | 0.5744 | 91.1527 |
Framework versions
- Transformers 4.37.0.dev0
- Pytorch 2.1.2+cu121
- Datasets 2.16.2.dev0
- Tokenizers 0.15.0