metadata
license: mit
base_model: microsoft/speecht5_tts
tags:
- generated_from_trainer
model-index:
- name: speecht5_tts-wolof-v0.2
results: []
speecht5_tts-wolof-v0.2
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: 0.3924
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: 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: 500
- num_epochs: 30
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.5083 | 0.9997 | 1908 | 0.4490 |
| 0.4789 | 2.0 | 3817 | 0.4399 |
| 0.4684 | 2.9997 | 5725 | 0.4297 |
| 0.4549 | 4.0 | 7634 | 0.4173 |
| 0.4448 | 4.9997 | 9542 | 0.4123 |
| 0.443 | 6.0 | 11451 | 0.4080 |
| 0.4368 | 6.9997 | 13359 | 0.4059 |
| 0.4351 | 8.0 | 15268 | 0.4030 |
| 0.4319 | 8.9997 | 17176 | 0.4027 |
| 0.4298 | 10.0 | 19085 | 0.4005 |
| 0.4286 | 10.9997 | 20993 | 0.3996 |
| 0.428 | 12.0 | 22902 | 0.3989 |
| 0.4251 | 12.9997 | 24810 | 0.3962 |
| 0.4257 | 14.0 | 26719 | 0.3971 |
| 0.4213 | 14.9997 | 28627 | 0.3956 |
| 0.4245 | 16.0 | 30536 | 0.3949 |
| 0.4186 | 16.9997 | 32444 | 0.3950 |
| 0.4213 | 18.0 | 34353 | 0.3948 |
| 0.4179 | 18.9997 | 36261 | 0.3943 |
| 0.4177 | 20.0 | 38170 | 0.3952 |
| 0.416 | 20.9997 | 40078 | 0.3932 |
| 0.4167 | 22.0 | 41987 | 0.3921 |
| 0.4148 | 22.9997 | 43895 | 0.3935 |
| 0.4133 | 24.0 | 45804 | 0.3938 |
| 0.4169 | 24.9997 | 47712 | 0.3924 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.4.0+cu121
- Datasets 3.2.0
- Tokenizers 0.19.1