<!-- 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-medium-ft-B16-E6-DS1K
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: 0.4190
- Wer: 15.3298
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: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 6
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 1.1484 | 1.0 | 63 | 0.4370 | 19.5187 |
| 0.1607 | 2.0 | 126 | 0.3946 | 16.2210 |
| 0.0553 | 3.0 | 189 | 0.3951 | 15.9537 |
| 0.0174 | 4.0 | 252 | 0.4193 | 16.5775 |
| 0.0072 | 5.0 | 315 | 0.4133 | 15.4189 |
| 0.0027 | 6.0 | 378 | 0.4190 | 15.3298 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cpu
- Datasets 2.12.0
- Tokenizers 0.13.3
