metadata
library_name: transformers
license: mit
base_model: microsoft/speecht5_tts
tags:
- generated_from_trainer
model-index:
- name: speech_T5_jahani_Finetune_Persian
results: []
speech_T5_jahani_Finetune_Persian
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: 0.4611
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: 3e-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: 8000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.5591 | 0.8711 | 1000 | 0.5155 |
| 0.5348 | 1.7422 | 2000 | 0.4937 |
| 0.5155 | 2.6132 | 3000 | 0.4782 |
| 0.5127 | 3.4843 | 4000 | 0.4735 |
| 0.4908 | 4.3554 | 5000 | 0.4690 |
| 0.4951 | 5.2265 | 6000 | 0.4654 |
| 0.4935 | 6.0976 | 7000 | 0.4617 |
| 0.4988 | 6.9686 | 8000 | 0.4611 |
Framework versions
- Transformers 4.48.3
- Pytorch 2.6.0+cu124
- Datasets 3.4.1
- Tokenizers 0.21.1