<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
openai/whisper-small
This model is a fine-tuned version of openai/whisper-small on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.3551
- Wer: 12.1862
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: 32
- eval_batch_size: 32
- 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 | Wer |
|---|---|---|---|---|
| 0.2646 | 0.12 | 500 | 0.3028 | 13.1170 |
| 0.2146 | 1.11 | 1000 | 0.2739 | 13.9555 |
| 0.2173 | 2.1 | 1500 | 0.2741 | 11.9619 |
| 0.0458 | 3.09 | 2000 | 0.2948 | 11.2066 |
| 0.0705 | 4.07 | 2500 | 0.3167 | 12.3373 |
| 0.0868 | 5.06 | 3000 | 0.3281 | 11.9397 |
| 0.0164 | 6.05 | 3500 | 0.3605 | 11.8886 |
| 0.0236 | 7.03 | 4000 | 0.3551 | 12.1862 |
Framework versions
- Transformers 4.27.0.dev0
- Pytorch 1.13.1+cu117
- Datasets 2.9.1.dev0
- Tokenizers 0.13.2
