|
--- |
|
license: mit |
|
base_model: microsoft/speecht5_tts |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- common_voice_13_0 |
|
model-index: |
|
- name: speecht5_tts_commonvoice_it_v2 |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# speecht5_tts_commonvoice_it_v2 |
|
|
|
This model is a fine-tuned version of [microsoft/speecht5_tts](https://huggingface.co/microsoft/speecht5_tts) on the common_voice_13_0 dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.5556 |
|
|
|
## 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-06 |
|
- train_batch_size: 32 |
|
- eval_batch_size: 16 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 2 |
|
- total_train_batch_size: 64 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_steps: 200 |
|
- training_steps: 2000 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:------:|:----:|:---------------:| |
|
| 0.9351 | 0.1894 | 100 | 0.8355 | |
|
| 0.8426 | 0.3788 | 200 | 0.7500 | |
|
| 0.8314 | 0.5682 | 300 | 0.7244 | |
|
| 0.7912 | 0.7576 | 400 | 0.7078 | |
|
| 0.778 | 0.9470 | 500 | 0.6908 | |
|
| 0.7205 | 1.1364 | 600 | 0.6744 | |
|
| 0.7272 | 1.3258 | 700 | 0.6469 | |
|
| 0.7394 | 1.5152 | 800 | 0.6176 | |
|
| 0.6816 | 1.7045 | 900 | 0.5874 | |
|
| 0.6653 | 1.8939 | 1000 | 0.5748 | |
|
| 0.658 | 2.0833 | 1100 | 0.5683 | |
|
| 0.628 | 2.2727 | 1200 | 0.5662 | |
|
| 0.6376 | 2.4621 | 1300 | 0.5632 | |
|
| 0.6232 | 2.6515 | 1400 | 0.5612 | |
|
| 0.625 | 2.8409 | 1500 | 0.5583 | |
|
| 0.63 | 3.0303 | 1600 | 0.5588 | |
|
| 0.6299 | 3.2197 | 1700 | 0.5567 | |
|
| 0.6332 | 3.4091 | 1800 | 0.5558 | |
|
| 0.6083 | 3.5985 | 1900 | 0.5551 | |
|
| 0.6161 | 3.7879 | 2000 | 0.5556 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.43.1 |
|
- Pytorch 2.2.0 |
|
- Datasets 3.0.1 |
|
- Tokenizers 0.19.1 |
|
|
