Whisper Medium Swedish
This model is a fine-tuned version of openai/whisper-medium on the Common Voice 11.0 dataset.
It achieves the following results on the evaluation set:
- Loss: 0.2594
- Wer: 13.9765
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: 4000
- mixed_precision_training: Native AMP
Training results
| Training Loss |
Epoch |
Step |
Validation Loss |
Wer |
| 0.0797 |
1.29 |
1000 |
0.2181 |
15.7215 |
| 0.0355 |
2.59 |
2000 |
0.2205 |
15.1408 |
| 0.0087 |
3.88 |
3000 |
0.2359 |
14.4444 |
| 0.0011 |
5.17 |
4000 |
0.2594 |
13.9765 |
Framework versions
- Transformers 4.25.0.dev0
- Pytorch 1.13.0+cu117
- Datasets 2.3.2
- Tokenizers 0.12.1
