<!-- 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. -->
whisper_small_tw8
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: 3.0869
- Wer: 1.2058
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: 6.25e-06
- train_batch_size: 16
- 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: 1200
- training_steps: 2400
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 2.6067 | 2.92 | 400 | 2.8624 | 2.0150 |
| 1.4795 | 5.84 | 800 | 2.2551 | 2.0746 |
| 0.6801 | 8.76 | 1200 | 2.4568 | 1.2906 |
| 0.1654 | 11.68 | 1600 | 2.7040 | 1.1842 |
| 0.0327 | 14.6 | 2000 | 2.9412 | 1.2688 |
| 0.0092 | 17.52 | 2400 | 3.0869 | 1.2058 |
Framework versions
- Transformers 4.34.0.dev0
- Pytorch 2.0.1+cu118
- Datasets 2.14.5
- Tokenizers 0.13.3
