SpeechT5-TTS-v1 for Thai
This model is a fine-tuned version of microsoft/speecht5_tts on the Edited Thai Common Voice dataset. It achieves the following results on the evaluation set:
- Loss: 0.5074
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
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- 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 |
|---|---|---|---|
| 0.5847 | 0.9794 | 1000 | 0.5360 |
| 0.5592 | 1.9589 | 2000 | 0.5158 |
| 0.5469 | 2.9383 | 3000 | 0.5103 |
| 0.5479 | 3.9177 | 4000 | 0.5074 |
Framework versions
- Transformers 4.52.0.dev0
- Pytorch 2.6.0+cu124
- Datasets 3.5.0
- Tokenizers 0.21.1