Wisper-Small-zh_test
This model is a fine-tuned version of openai/whisper-small on the Common Voice 16.1 dataset. It achieves the following results on the evaluation set:
- Loss: 0.3867
- Cer: 21.2804
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: 2000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Cer | Validation Loss |
|---|---|---|---|---|
| 2.9388 | 0.6623 | 500 | 102.7090 | 2.9818 |
| 0.5492 | 1.3245 | 1000 | 33.3994 | 0.6020 |
| 0.352 | 1.9868 | 1500 | 0.4214 | 23.2990 |
| 0.2243 | 2.6490 | 2000 | 0.3867 | 21.2804 |
Framework versions
- Transformers 4.42.0.dev0
- Pytorch 2.1.2
- Datasets 2.20.0
- Tokenizers 0.19.1
- Downloads last month
- 1
Model tree for jun-han/whisper-small
Base model
openai/whisper-small