whispertiny_kokoro_styletts_25sil_v1_4
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: 0.1027
- Wer: 7.6895
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: 10000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.1044 | 0.2253 | 1000 | 0.1460 | 15.4797 |
| 0.0434 | 0.4507 | 2000 | 0.1159 | 11.1966 |
| 0.0569 | 0.6760 | 3000 | 0.1189 | 10.2551 |
| 0.0369 | 0.9013 | 4000 | 0.0925 | 8.2285 |
| 0.0218 | 1.1266 | 5000 | 0.1095 | 9.6658 |
| 0.0238 | 1.3520 | 6000 | 0.1004 | 8.3579 |
| 0.0202 | 1.5773 | 7000 | 0.0903 | 7.3158 |
| 0.0152 | 1.8026 | 8000 | 0.1019 | 7.8620 |
| 0.0114 | 2.0279 | 9000 | 0.0969 | 7.1218 |
| 0.0092 | 2.2533 | 10000 | 0.1027 | 7.6895 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 4.0.0
- Tokenizers 0.19.1