metadata
license: mit
base_model: microsoft/speecht5_tts
tags:
- generated_from_trainer
model-index:
- name: speecht5_tts
results: []
speecht5_tts
This model is a fine-tuned version of microsoft/speecht5_tts on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.6698
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: 4
- 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: 500
- training_steps: 15000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| No log | 11.36 | 250 | 0.7543 |
| 0.7974 | 22.73 | 500 | 0.5421 |
| 0.7974 | 34.09 | 750 | 0.5298 |
| 0.5132 | 45.45 | 1000 | 0.5404 |
| 0.5132 | 56.82 | 1250 | 0.5513 |
| 0.4589 | 68.18 | 1500 | 0.5664 |
| 0.4589 | 79.55 | 1750 | 0.5688 |
| 0.4362 | 90.91 | 2000 | 0.6007 |
| 0.4362 | 102.27 | 2250 | 0.5819 |
| 0.4264 | 113.64 | 2500 | 0.5813 |
| 0.4264 | 125.0 | 2750 | 0.6025 |
| 0.4136 | 136.36 | 3000 | 0.6001 |
| 0.4136 | 147.73 | 3250 | 0.5854 |
| 0.4058 | 159.09 | 3500 | 0.6119 |
| 0.4058 | 170.45 | 3750 | 0.6194 |
| 0.3914 | 181.82 | 4000 | 0.6248 |
| 0.3914 | 193.18 | 4250 | 0.6185 |
| 0.3892 | 204.55 | 4500 | 0.6116 |
| 0.3892 | 215.91 | 4750 | 0.6208 |
| 0.3807 | 227.27 | 5000 | 0.6275 |
| 0.3807 | 238.64 | 5250 | 0.6214 |
| 0.378 | 250.0 | 5500 | 0.6209 |
| 0.378 | 261.36 | 5750 | 0.6398 |
| 0.3677 | 272.73 | 6000 | 0.6272 |
| 0.3677 | 284.09 | 6250 | 0.6274 |
| 0.3662 | 295.45 | 6500 | 0.6571 |
| 0.3662 | 306.82 | 6750 | 0.6337 |
| 0.3593 | 318.18 | 7000 | 0.6456 |
| 0.3593 | 329.55 | 7250 | 0.6321 |
| 0.3587 | 340.91 | 7500 | 0.6360 |
| 0.3587 | 352.27 | 7750 | 0.6504 |
| 0.3556 | 363.64 | 8000 | 0.6478 |
| 0.3556 | 375.0 | 8250 | 0.6516 |
| 0.3554 | 386.36 | 8500 | 0.6398 |
| 0.3554 | 397.73 | 8750 | 0.6524 |
| 0.3489 | 409.09 | 9000 | 0.6484 |
| 0.3489 | 420.45 | 9250 | 0.6418 |
| 0.3513 | 431.82 | 9500 | 0.6421 |
| 0.3513 | 443.18 | 9750 | 0.6460 |
| 0.3464 | 454.55 | 10000 | 0.6586 |
| 0.3464 | 465.91 | 10250 | 0.6613 |
| 0.3456 | 477.27 | 10500 | 0.6511 |
| 0.3456 | 488.64 | 10750 | 0.6474 |
| 0.3405 | 500.0 | 11000 | 0.6402 |
| 0.3405 | 511.36 | 11250 | 0.6541 |
| 0.3436 | 522.73 | 11500 | 0.6596 |
| 0.3436 | 534.09 | 11750 | 0.6631 |
| 0.3385 | 545.45 | 12000 | 0.6574 |
| 0.3385 | 556.82 | 12250 | 0.6516 |
| 0.3405 | 568.18 | 12500 | 0.6675 |
| 0.3405 | 579.55 | 12750 | 0.6617 |
| 0.3403 | 590.91 | 13000 | 0.6573 |
| 0.3403 | 602.27 | 13250 | 0.6595 |
| 0.336 | 613.64 | 13500 | 0.6625 |
| 0.336 | 625.0 | 13750 | 0.6601 |
| 0.3376 | 636.36 | 14000 | 0.6655 |
| 0.3376 | 647.73 | 14250 | 0.6695 |
| 0.3357 | 659.09 | 14500 | 0.6623 |
| 0.3357 | 670.45 | 14750 | 0.6576 |
| 0.3348 | 681.82 | 15000 | 0.6698 |
Framework versions
- Transformers 4.36.0.dev0
- Pytorch 2.1.0+cu121
- Datasets 2.15.0
- Tokenizers 0.14.1