speecht5_mehdi_as_Multi_token
This model is a fine-tuned version of microsoft/speecht5_tts on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.4877
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: 0.0001
- train_batch_size: 4
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 32
- optimizer: Use 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: 100
- training_steps: 1000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.6154 | 3.5556 | 100 | 0.5500 |
| 0.5594 | 7.1111 | 200 | 0.5302 |
| 0.5378 | 10.6667 | 300 | 0.5179 |
| 0.5086 | 14.2222 | 400 | 0.4924 |
| 0.4923 | 17.7778 | 500 | 0.4855 |
| 0.4871 | 21.3333 | 600 | 0.4844 |
| 0.4791 | 24.8889 | 700 | 0.4809 |
| 0.4724 | 28.4444 | 800 | 0.4793 |
| 0.4609 | 32.0 | 900 | 0.4837 |
| 0.4583 | 35.5556 | 1000 | 0.4877 |
Framework versions
- Transformers 4.46.3
- Pytorch 2.5.1+cu121
- Datasets 3.1.0
- Tokenizers 0.20.3