metadata
library_name: transformers
license: mit
base_model: microsoft/speecht5_tts
tags:
- generated_from_trainer
datasets:
- common_voice_17_0
model-index:
- name: SpeechT5-Hausa-7
results: []
SpeechT5-Hausa-7
This model is a fine-tuned version of microsoft/speecht5_tts on the common_voice_17_0 dataset. It achieves the following results on the evaluation set:
- Loss: 0.4667
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: 2000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.6236 | 1.8476 | 100 | 0.5467 |
| 0.5606 | 3.6952 | 200 | 0.5343 |
| 0.537 | 5.5427 | 300 | 0.4949 |
| 0.5195 | 7.3903 | 400 | 0.4839 |
| 0.511 | 9.2379 | 500 | 0.4760 |
| 0.5044 | 11.0855 | 600 | 0.4693 |
| 0.4944 | 12.9330 | 700 | 0.4693 |
| 0.4912 | 14.7806 | 800 | 0.4843 |
| 0.4792 | 16.6282 | 900 | 0.4737 |
| 0.4742 | 18.4758 | 1000 | 0.4659 |
| 0.4727 | 20.3233 | 1100 | 0.4659 |
| 0.4584 | 22.1709 | 1200 | 0.4637 |
| 0.4604 | 24.0185 | 1300 | 0.4618 |
| 0.4556 | 25.8661 | 1400 | 0.4637 |
| 0.4549 | 27.7136 | 1500 | 0.4606 |
| 0.4514 | 29.5612 | 1600 | 0.4629 |
| 0.4476 | 31.4088 | 1700 | 0.4611 |
| 0.4426 | 33.2564 | 1800 | 0.4636 |
| 0.4472 | 35.1039 | 1900 | 0.4680 |
| 0.4353 | 36.9515 | 2000 | 0.4667 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.4.1+cu121
- Datasets 3.0.0
- Tokenizers 0.19.1