|
--- |
|
library_name: transformers |
|
license: mit |
|
base_model: microsoft/speecht5_tts |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: speecht5_finetuned_commonvoice_dv |
|
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_finetuned_commonvoice_dv |
|
|
|
This model is a fine-tuned version of [microsoft/speecht5_tts](https://huggingface.co/microsoft/speecht5_tts) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.4334 |
|
|
|
## 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: 32 |
|
- eval_batch_size: 16 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 12 |
|
- total_train_batch_size: 384 |
|
- 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: 4000 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:--------:|:----:|:---------------:| |
|
| 10.3761 | 3.2222 | 100 | 0.7551 | |
|
| 8.3203 | 6.4444 | 200 | 0.5722 | |
|
| 7.3507 | 9.6667 | 300 | 0.5280 | |
|
| 6.9851 | 12.8889 | 400 | 0.5115 | |
|
| 6.6688 | 16.1270 | 500 | 0.4952 | |
|
| 6.479 | 19.3492 | 600 | 0.4871 | |
|
| 6.4798 | 22.5714 | 700 | 0.4771 | |
|
| 6.2714 | 25.7937 | 800 | 0.4759 | |
|
| 6.2132 | 29.0317 | 900 | 0.4700 | |
|
| 6.1966 | 32.2540 | 1000 | 0.4652 | |
|
| 6.1389 | 35.4762 | 1100 | 0.4638 | |
|
| 6.0647 | 38.6984 | 1200 | 0.4603 | |
|
| 6.032 | 41.9206 | 1300 | 0.4602 | |
|
| 6.0107 | 45.1587 | 1400 | 0.4552 | |
|
| 5.9762 | 48.3810 | 1500 | 0.4522 | |
|
| 6.0347 | 51.6032 | 1600 | 0.4507 | |
|
| 5.9424 | 54.8254 | 1700 | 0.4508 | |
|
| 5.9278 | 58.0635 | 1800 | 0.4522 | |
|
| 5.9332 | 61.2857 | 1900 | 0.4473 | |
|
| 5.9201 | 64.5079 | 2000 | 0.4443 | |
|
| 5.8812 | 67.7302 | 2100 | 0.4439 | |
|
| 5.8007 | 70.9524 | 2200 | 0.4426 | |
|
| 5.8262 | 74.1905 | 2300 | 0.4409 | |
|
| 5.8343 | 77.4127 | 2400 | 0.4404 | |
|
| 5.8536 | 80.6349 | 2500 | 0.4408 | |
|
| 5.7672 | 83.8571 | 2600 | 0.4381 | |
|
| 5.757 | 87.0952 | 2700 | 0.4381 | |
|
| 5.7981 | 90.3175 | 2800 | 0.4366 | |
|
| 5.8329 | 93.5397 | 2900 | 0.4371 | |
|
| 5.7738 | 96.7619 | 3000 | 0.4365 | |
|
| 5.7674 | 99.9841 | 3100 | 0.4370 | |
|
| 5.7987 | 103.2222 | 3200 | 0.4356 | |
|
| 5.6883 | 106.4444 | 3300 | 0.4351 | |
|
| 5.7883 | 109.6667 | 3400 | 0.4374 | |
|
| 5.7269 | 112.8889 | 3500 | 0.4345 | |
|
| 5.723 | 116.1270 | 3600 | 0.4336 | |
|
| 5.7776 | 119.3492 | 3700 | 0.4354 | |
|
| 5.724 | 122.5714 | 3800 | 0.4342 | |
|
| 5.7235 | 125.7937 | 3900 | 0.4334 | |
|
| 5.7067 | 129.0317 | 4000 | 0.4334 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.48.0.dev0 |
|
- Pytorch 2.5.1+cu121 |
|
- Datasets 3.2.0 |
|
- Tokenizers 0.21.0 |
|
|
