|
--- |
|
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: 0.5076 |
|
|
|
## 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: 4 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_steps: 1000 |
|
- num_epochs: 3 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:------:|:-----:|:---------------:| |
|
| 0.9213 | 0.0994 | 500 | 0.7823 | |
|
| 0.8356 | 0.1987 | 1000 | 0.7026 | |
|
| 0.6804 | 0.2981 | 1500 | 0.6003 | |
|
| 0.6518 | 0.3975 | 2000 | 0.5751 | |
|
| 0.6242 | 0.4968 | 2500 | 0.5594 | |
|
| 0.6237 | 0.5962 | 3000 | 0.5514 | |
|
| 0.6122 | 0.6955 | 3500 | 0.5414 | |
|
| 0.597 | 0.7949 | 4000 | 0.5335 | |
|
| 0.5909 | 0.8943 | 4500 | 0.5322 | |
|
| 0.6009 | 0.9936 | 5000 | 0.5283 | |
|
| 0.6086 | 1.0930 | 5500 | 0.5258 | |
|
| 0.5812 | 1.1924 | 6000 | 0.5209 | |
|
| 0.5868 | 1.2917 | 6500 | 0.5191 | |
|
| 0.5689 | 1.3911 | 7000 | 0.5177 | |
|
| 0.5777 | 1.4905 | 7500 | 0.5182 | |
|
| 0.577 | 1.5898 | 8000 | 0.5169 | |
|
| 0.5594 | 1.6892 | 8500 | 0.5150 | |
|
| 0.5728 | 1.7886 | 9000 | 0.5144 | |
|
| 0.571 | 1.8879 | 9500 | 0.5125 | |
|
| 0.5739 | 1.9873 | 10000 | 0.5116 | |
|
| 0.5819 | 2.0866 | 10500 | 0.5102 | |
|
| 0.5633 | 2.1860 | 11000 | 0.5102 | |
|
| 0.5635 | 2.2854 | 11500 | 0.5093 | |
|
| 0.5809 | 2.3847 | 12000 | 0.5094 | |
|
| 0.5647 | 2.4841 | 12500 | 0.5086 | |
|
| 0.5593 | 2.5835 | 13000 | 0.5065 | |
|
| 0.5639 | 2.6828 | 13500 | 0.5077 | |
|
| 0.5511 | 2.7822 | 14000 | 0.5073 | |
|
| 0.5534 | 2.8816 | 14500 | 0.5071 | |
|
| 0.5532 | 2.9809 | 15000 | 0.5076 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.43.1 |
|
- Pytorch 2.4.1+cu121 |
|
- Datasets 3.0.0 |
|
- Tokenizers 0.19.1 |
|
|
