metadata
license: mit
base_model: microsoft/speecht5_tts
tags:
- generated_from_trainer
model-index:
- name: speecht5_jenny_500sample
results: []
speecht5_jenny_500sample
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: 0.4246
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: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 4
- 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: 100
- training_steps: 500
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.5771 | 1.7699 | 50 | 0.4779 |
| 0.5203 | 3.5398 | 100 | 0.4469 |
| 0.4659 | 5.3097 | 150 | 0.4358 |
| 0.4456 | 7.0796 | 200 | 0.4305 |
| 0.4417 | 8.8496 | 250 | 0.4217 |
| 0.4299 | 10.6195 | 300 | 0.4218 |
| 0.4216 | 12.3894 | 350 | 0.4168 |
| 0.419 | 14.1593 | 400 | 0.4291 |
| 0.4132 | 15.9292 | 450 | 0.4273 |
| 0.4203 | 17.6991 | 500 | 0.4246 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.1.0+cu118
- Datasets 2.20.0
- Tokenizers 0.19.1