Swedish-Fine-Tuned-Whisper-Model
This model is a fine-tuned version of openai/whisper-tiny on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.0729
- Wer: 156.0742
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: 8
- 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: 100
- training_steps: 1000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.4797 | 1.6 | 200 | 0.9948 | 119.6292 |
| 0.1806 | 3.2 | 400 | 0.9994 | 151.1509 |
| 0.1107 | 4.8 | 600 | 1.0260 | 139.2583 |
| 0.0582 | 6.4 | 800 | 1.0565 | 129.5396 |
| 0.0416 | 8.0 | 1000 | 1.0729 | 156.0742 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.13.1+cu116
- Datasets 2.10.1
- Tokenizers 0.13.2