metadata
library_name: transformers
license: mit
base_model: microsoft/speecht5_tts
tags:
- generated_from_trainer
datasets:
- voxpopuli
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 dataset. It achieves the following results on the evaluation set:
- Loss: 0.4599
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: 16
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- 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: 500
- training_steps: 2000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.679 | 4.0 | 200 | 0.5742 |
| 0.5413 | 8.0 | 400 | 0.4974 |
| 0.5076 | 12.0 | 600 | 0.4805 |
| 0.5097 | 16.0 | 800 | 0.4743 |
| 0.4837 | 20.0 | 1000 | 0.4655 |
| 0.4913 | 24.0 | 1200 | 0.4639 |
| 0.4777 | 28.0 | 1400 | 0.4632 |
| 0.484 | 32.0 | 1600 | 0.4623 |
| 0.4788 | 36.0 | 1800 | 0.4608 |
| 0.4711 | 40.0 | 2000 | 0.4599 |
Framework versions
- Transformers 4.54.1
- Pytorch 2.7.1+cu126
- Datasets 3.6.0
- Tokenizers 0.21.4