metadata
library_name: transformers
license: mit
base_model: microsoft/speecht5_tts
tags:
- generated_from_trainer
- text-to-speech
datasets:
- voxpopuli/fi
model-index:
- name: speecht5_finetuned_voxpopuli_fi
results: []
speecht5_finetuned_voxpopuli_fi
This model is a fine-tuned version of microsoft/speecht5_tts on the voxpopuli/fi dataset. It achieves the following results on the evaluation set:
- Loss: 0.4489
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: 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: 250
- training_steps: 4000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.5539 | 1.4577 | 500 | 0.4861 |
| 0.5132 | 2.9155 | 1000 | 0.4683 |
| 0.5069 | 4.3732 | 1500 | 0.4607 |
| 0.4886 | 5.8309 | 2000 | 0.4555 |
| 0.495 | 7.2886 | 2500 | 0.4539 |
| 0.4846 | 8.7464 | 3000 | 0.4504 |
| 0.4843 | 10.2041 | 3500 | 0.4495 |
| 0.4932 | 11.6618 | 4000 | 0.4489 |
Framework versions
- Transformers 4.46.2
- Pytorch 2.5.1+cu121
- Datasets 3.1.0
- Tokenizers 0.20.3