|
|
--- |
|
|
library_name: transformers |
|
|
license: mit |
|
|
base_model: microsoft/speecht5_tts |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
model-index: |
|
|
- name: EGPABG |
|
|
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. --> |
|
|
|
|
|
# EGPABG |
|
|
|
|
|
This model is a fine-tuned version of [microsoft/speecht5_tts](https://huggingface.co/microsoft/speecht5_tts) on an unknown dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 0.5020 |
|
|
|
|
|
## 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: 0.0001 |
|
|
- train_batch_size: 4 |
|
|
- eval_batch_size: 2 |
|
|
- seed: 42 |
|
|
- gradient_accumulation_steps: 8 |
|
|
- 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 |
|
|
- training_steps: 3000 |
|
|
- mixed_precision_training: Native AMP |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|
|:-------------:|:------:|:----:|:---------------:| |
|
|
| 0.6281 | 0.2540 | 100 | 0.5862 | |
|
|
| 0.6048 | 0.5079 | 200 | 0.5734 | |
|
|
| 0.5792 | 0.7619 | 300 | 0.5444 | |
|
|
| 0.57 | 1.0159 | 400 | 0.5418 | |
|
|
| 0.5657 | 1.2698 | 500 | 0.5281 | |
|
|
| 0.5625 | 1.5238 | 600 | 0.5301 | |
|
|
| 0.5585 | 1.7778 | 700 | 0.5245 | |
|
|
| 0.5559 | 2.0317 | 800 | 0.5263 | |
|
|
| 0.5572 | 2.2857 | 900 | 0.5263 | |
|
|
| 0.5527 | 2.5397 | 1000 | 0.5161 | |
|
|
| 0.5524 | 2.7937 | 1100 | 0.5190 | |
|
|
| 0.544 | 3.0476 | 1200 | 0.5154 | |
|
|
| 0.542 | 3.3016 | 1300 | 0.5203 | |
|
|
| 0.5425 | 3.5556 | 1400 | 0.5163 | |
|
|
| 0.5413 | 3.8095 | 1500 | 0.5099 | |
|
|
| 0.5328 | 4.0635 | 1600 | 0.5149 | |
|
|
| 0.5406 | 4.3175 | 1700 | 0.5107 | |
|
|
| 0.5347 | 4.5714 | 1800 | 0.5079 | |
|
|
| 0.5362 | 4.8254 | 1900 | 0.5068 | |
|
|
| 0.5355 | 5.0794 | 2000 | 0.5050 | |
|
|
| 0.5312 | 5.3333 | 2100 | 0.5061 | |
|
|
| 0.5282 | 5.5873 | 2200 | 0.5081 | |
|
|
| 0.5298 | 5.8413 | 2300 | 0.5029 | |
|
|
| 0.5288 | 6.0952 | 2400 | 0.5028 | |
|
|
| 0.5371 | 6.3492 | 2500 | 0.5023 | |
|
|
| 0.525 | 6.6032 | 2600 | 0.5022 | |
|
|
| 0.5281 | 6.8571 | 2700 | 0.5039 | |
|
|
| 0.5223 | 7.1111 | 2800 | 0.5023 | |
|
|
| 0.5231 | 7.3651 | 2900 | 0.5022 | |
|
|
| 0.5243 | 7.6190 | 3000 | 0.5020 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.44.2 |
|
|
- Pytorch 2.4.1+cu121 |
|
|
- Datasets 3.0.0 |
|
|
- Tokenizers 0.19.1 |
|
|
|
