Whisper Small Slovak test on Czech
This model is a fine-tuned version of mikr/whisper-small-cs-cv11 on the mozilla-foundation/common_voice_11_0 sk dataset. It achieves the following results on the evaluation set:
- Loss: 0.7223
- Wer: 35.4355
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: 64
- 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: 5000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.001 | 21.0 | 1000 | 0.6507 | 37.3275 |
| 0.0003 | 42.01 | 2000 | 0.6954 | 36.1138 |
| 0.0002 | 63.01 | 3000 | 0.7223 | 35.4355 |
| 0.0001 | 85.0 | 4000 | 0.7388 | 35.5902 |
| 0.0001 | 106.0 | 5000 | 0.7465 | 35.6735 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu117
- Datasets 2.7.1.dev0
- Tokenizers 0.13.2