metadata
license: mit
tags:
- generated_from_trainer
datasets:
- voxpopuli
model-index:
- name: speecht5_finetuned_facebook_voxpopuli_french
results: []
pipeline_tag: text-to-speech
speecht5_finetuned_facebook_voxpopuli_french
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: 0.4378
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: 5e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- num_epochs: 20
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.4873 | 1.0 | 1584 | 0.4617 |
| 0.4689 | 2.0 | 3168 | 0.4529 |
| 0.4658 | 3.0 | 4752 | 0.4516 |
| 0.4584 | 4.0 | 6336 | 0.4464 |
| 0.4608 | 5.0 | 7920 | 0.4451 |
| 0.4558 | 6.0 | 9504 | 0.4427 |
| 0.4504 | 7.0 | 11088 | 0.4426 |
| 0.4528 | 8.0 | 12672 | 0.4409 |
| 0.4515 | 9.0 | 14256 | 0.4397 |
| 0.4489 | 10.0 | 15840 | 0.4409 |
| 0.4452 | 11.0 | 17424 | 0.4393 |
| 0.4403 | 12.0 | 19008 | 0.4394 |
| 0.4419 | 13.0 | 20592 | 0.4392 |
| 0.441 | 14.0 | 22176 | 0.4378 |
| 0.4418 | 15.0 | 23760 | 0.4376 |
| 0.4373 | 16.0 | 25344 | 0.4382 |
| 0.4405 | 17.0 | 26928 | 0.4367 |
| 0.4377 | 18.0 | 28512 | 0.4374 |
| 0.4351 | 19.0 | 30096 | 0.4376 |
| 0.4371 | 20.0 | 31680 | 0.4378 |
Framework versions
- Transformers 4.30.0.dev0
- Pytorch 2.0.1+cu117
- Datasets 2.13.1
- Tokenizers 0.13.3