metadata
library_name: transformers
license: mit
base_model: microsoft/speecht5_tts
tags:
- generated_from_trainer
model-index:
- name: speecht5_finetuned_librispeech_polish_epo6_batch2_gas3
results: []
speecht5_finetuned_librispeech_polish_epo6_batch2_gas3
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: 0.3676
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: 2
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 3
- total_train_batch_size: 6
- 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
- num_epochs: 6
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.4801 | 0.3836 | 100 | 0.4534 |
| 1.4511 | 0.7673 | 200 | 0.4228 |
| 1.3122 | 1.1496 | 300 | 0.4199 |
| 1.285 | 1.5332 | 400 | 0.3972 |
| 1.268 | 1.9169 | 500 | 0.3930 |
| 1.2609 | 2.2992 | 600 | 0.3869 |
| 1.2335 | 2.6829 | 700 | 0.3861 |
| 1.2122 | 3.0652 | 800 | 0.3843 |
| 1.2336 | 3.4488 | 900 | 0.3801 |
| 1.2077 | 3.8325 | 1000 | 0.3783 |
| 1.2312 | 4.2148 | 1100 | 0.3743 |
| 1.1972 | 4.5985 | 1200 | 0.3710 |
| 1.1902 | 4.9821 | 1300 | 0.3697 |
| 1.1841 | 5.3645 | 1400 | 0.3697 |
| 1.1713 | 5.7481 | 1500 | 0.3676 |
Framework versions
- Transformers 4.47.1
- Pytorch 2.5.1+cu121
- Datasets 3.2.0
- Tokenizers 0.21.0