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: 0.7779
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 | 31.25 | 250 | 0.8090 |
| 0.7576 | 62.5 | 500 | 0.6545 |
| 0.7576 | 93.75 | 750 | 0.6266 |
| 0.4719 | 125.0 | 1000 | 0.6372 |
| 0.4719 | 156.25 | 1250 | 0.6317 |
| 0.403 | 187.5 | 1500 | 0.6457 |
| 0.403 | 218.75 | 1750 | 0.6571 |
| 0.374 | 250.0 | 2000 | 0.6805 |
| 0.374 | 281.25 | 2250 | 0.6814 |
| 0.3564 | 312.5 | 2500 | 0.6814 |
| 0.3564 | 343.75 | 2750 | 0.6956 |
| 0.342 | 375.0 | 3000 | 0.6769 |
| 0.342 | 406.25 | 3250 | 0.7000 |
| 0.3339 | 437.5 | 3500 | 0.7052 |
| 0.3339 | 468.75 | 3750 | 0.7104 |
| 0.3205 | 500.0 | 4000 | 0.7095 |
| 0.3205 | 531.25 | 4250 | 0.7120 |
| 0.3112 | 562.5 | 4500 | 0.7194 |
| 0.3112 | 593.75 | 4750 | 0.7363 |
| 0.3045 | 625.0 | 5000 | 0.7252 |
| 0.3045 | 656.25 | 5250 | 0.7003 |
| 0.3067 | 687.5 | 5500 | 0.7176 |
| 0.3067 | 718.75 | 5750 | 0.7513 |
| 0.2938 | 750.0 | 6000 | 0.7403 |
| 0.2938 | 781.25 | 6250 | 0.7180 |
| 0.2894 | 812.5 | 6500 | 0.7569 |
| 0.2894 | 843.75 | 6750 | 0.7398 |
| 0.2886 | 875.0 | 7000 | 0.7384 |
| 0.2886 | 906.25 | 7250 | 0.7363 |
| 0.2877 | 937.5 | 7500 | 0.7550 |
| 0.2877 | 968.75 | 7750 | 0.7350 |
| 0.2822 | 1000.0 | 8000 | 0.8012 |
| 0.2822 | 1031.25 | 8250 | 0.7509 |
| 0.2808 | 1062.5 | 8500 | 0.7544 |
| 0.2808 | 1093.75 | 8750 | 0.7851 |
| 0.2746 | 1125.0 | 9000 | 0.7283 |
| 0.2746 | 1156.25 | 9250 | 0.7559 |
| 0.27 | 1187.5 | 9500 | 0.7449 |
| 0.27 | 1218.75 | 9750 | 0.7496 |
| 0.2693 | 1250.0 | 10000 | 0.7375 |
| 0.2693 | 1281.25 | 10250 | 0.7723 |
| 0.2674 | 1312.5 | 10500 | 0.7632 |
| 0.2674 | 1343.75 | 10750 | 0.7825 |
| 0.2678 | 1375.0 | 11000 | 0.7645 |
| 0.2678 | 1406.25 | 11250 | 0.7502 |
| 0.269 | 1437.5 | 11500 | 0.7762 |
| 0.269 | 1468.75 | 11750 | 0.7500 |
| 0.2642 | 1500.0 | 12000 | 0.7503 |
| 0.2642 | 1531.25 | 12250 | 0.7460 |
| 0.2618 | 1562.5 | 12500 | 0.7512 |
| 0.2618 | 1593.75 | 12750 | 0.7889 |
| 0.2642 | 1625.0 | 13000 | 0.7578 |
| 0.2642 | 1656.25 | 13250 | 0.7678 |
| 0.2628 | 1687.5 | 13500 | 0.7638 |
| 0.2628 | 1718.75 | 13750 | 0.7721 |
| 0.2612 | 1750.0 | 14000 | 0.7531 |
| 0.2612 | 1781.25 | 14250 | 0.7924 |
| 0.2555 | 1812.5 | 14500 | 0.7882 |
| 0.2555 | 1843.75 | 14750 | 0.7841 |
| 0.2596 | 1875.0 | 15000 | 0.7779 |
Framework versions
- Transformers 4.36.0.dev0
- Pytorch 2.1.0+cu121
- Datasets 2.15.0
- Tokenizers 0.14.1