metadata
license: mit
base_model: microsoft/speecht5_tts
tags:
- generated_from_trainer
model-index:
- name: UAE1k
results: []
UAE1k
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: 0.4901
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: 2
- 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: 1000
- training_steps: 2500
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.5308 | 4.08 | 200 | 0.4912 |
| 0.5091 | 8.16 | 400 | 0.4925 |
| 0.4987 | 12.24 | 600 | 0.4895 |
| 0.5085 | 16.33 | 800 | 0.4896 |
| 0.5055 | 20.41 | 1000 | 0.4894 |
| 0.4941 | 24.49 | 1200 | 0.4972 |
| 0.4833 | 28.57 | 1400 | 0.4888 |
| 0.4963 | 32.65 | 1600 | 0.4916 |
| 0.4803 | 36.73 | 1800 | 0.4906 |
| 0.4918 | 40.82 | 2000 | 0.4904 |
| 0.4765 | 44.9 | 2200 | 0.4886 |
| 0.4782 | 48.98 | 2400 | 0.4901 |
Framework versions
- Transformers 4.39.3
- Pytorch 2.3.0+cu118
- Datasets 3.0.0
- Tokenizers 0.15.2