metadata
library_name: transformers
license: mit
base_model: microsoft/speecht5_tts
tags:
- generated_from_trainer
model-index:
- name: speecht5_mehdi_as_1
results: []
speecht5_mehdi_as_1
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: 0.5176
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: 1500
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.611 | 3.5556 | 100 | 0.5611 |
| 0.55 | 7.1111 | 200 | 0.5361 |
| 0.5435 | 10.6667 | 300 | 0.5158 |
| 0.5081 | 14.2222 | 400 | 0.4987 |
| 0.4918 | 17.7778 | 500 | 0.5124 |
| 0.4851 | 21.3333 | 600 | 0.4984 |
| 0.4783 | 24.8889 | 700 | 0.5027 |
| 0.4721 | 28.4444 | 800 | 0.4964 |
| 0.4595 | 32.0 | 900 | 0.5092 |
| 0.4524 | 35.5556 | 1000 | 0.5169 |
| 0.4528 | 39.1111 | 1100 | 0.5130 |
| 0.4423 | 42.6667 | 1200 | 0.5114 |
| 0.4401 | 46.2222 | 1300 | 0.5175 |
| 0.439 | 49.7778 | 1400 | 0.5202 |
| 0.4357 | 53.3333 | 1500 | 0.5176 |
Framework versions
- Transformers 4.46.3
- Pytorch 2.5.1+cu121
- Tokenizers 0.20.3