metadata
license: mit
base_model: microsoft/speecht5_tts
tags:
- generated_from_trainer
model-index:
- name: ceb_b128_le4_s4000
results: []
ceb_b128_le4_s4000
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: 0.3981
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: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 2000
- training_steps: 4000
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
0.4433 | 39.6040 | 500 | 0.4056 |
0.4142 | 79.2079 | 1000 | 0.3953 |
0.3972 | 118.8119 | 1500 | 0.3951 |
0.3806 | 158.4158 | 2000 | 0.3970 |
0.3653 | 198.0198 | 2500 | 0.3960 |
0.3566 | 237.6238 | 3000 | 0.4034 |
0.349 | 277.2277 | 3500 | 0.3991 |
0.3466 | 316.8317 | 4000 | 0.3981 |
Framework versions
- Transformers 4.41.0.dev0
- Pytorch 2.2.1+cu121
- Datasets 2.19.0
- Tokenizers 0.19.1