Whisper Medium fine-tuned for Tatar language
This model is a fine-tuned version of openai/whisper-small on the Common Voice 16.1 dataset. It achieves the following results on the evaluation set:
- Loss: 0.2809
- Wer: 34.8445
Training and evaluation data
Training data was taken from Common Voice 16.1 dataset
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 250
- training_steps: 2000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.1399 | 1.2293 | 1000 | 0.3081 | 38.2040 |
| 0.0639 | 2.4585 | 2000 | 0.2809 | 34.8445 |
Framework versions
- Transformers 4.41.0
- Pytorch 2.1.2
- Datasets 2.19.1
- Tokenizers 0.19.1