metadata
library_name: transformers
license: mit
base_model: microsoft/speecht5_tts
tags:
- generated_from_trainer
model-index:
- name: TTS_FineTune_French
results: []
TTS_FineTune_French
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: 0.4810
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: 600
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 4.3698 | 16.6957 | 100 | 0.5334 |
| 3.9064 | 33.3478 | 200 | 0.5207 |
| 3.6468 | 50.0 | 300 | 0.4919 |
| 3.6001 | 66.6957 | 400 | 0.4892 |
| 3.4848 | 83.3478 | 500 | 0.4813 |
| 3.3729 | 100.0 | 600 | 0.4810 |
Framework versions
- Transformers 4.48.0.dev0
- Pytorch 2.5.1+cu121
- Datasets 3.2.0
- Tokenizers 0.21.0