openai/whisper-medium
This model is a fine-tuned version of openai/whisper-medium on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1711
- Wer: 10.1446
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-06
- train_batch_size: 32
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 200
- training_steps: 5000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.111 | 0.2 | 1000 | 0.1786 | 11.0063 |
| 0.0961 | 1.16 | 2000 | 0.1719 | 10.5906 |
| 0.0732 | 2.12 | 3000 | 0.1743 | 10.3268 |
| 0.0742 | 3.08 | 4000 | 0.1715 | 10.2262 |
| 0.0692 | 4.03 | 5000 | 0.1711 | 10.1446 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu117
- Datasets 2.7.1.dev0
- Tokenizers 0.13.2
- Downloads last month
- 1