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-9
results: []
SpeechT5-Hausa-9
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: 0.4650
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 |
|---|---|---|---|
| 0.6131 | 1.8476 | 100 | 0.5543 |
| 0.5611 | 3.6952 | 200 | 0.5054 |
| 0.5362 | 5.5427 | 300 | 0.4950 |
| 0.5222 | 7.3903 | 400 | 0.4837 |
| 0.5067 | 9.2379 | 500 | 0.4770 |
| 0.5028 | 11.0855 | 600 | 0.4733 |
| 0.4926 | 12.9330 | 700 | 0.4685 |
| 0.4833 | 14.7806 | 800 | 0.4640 |
| 0.4776 | 16.6282 | 900 | 0.4606 |
| 0.4769 | 18.4758 | 1000 | 0.4582 |
| 0.4715 | 20.3233 | 1100 | 0.4651 |
| 0.462 | 22.1709 | 1200 | 0.4612 |
| 0.4579 | 24.0185 | 1300 | 0.4616 |
| 0.455 | 25.8661 | 1400 | 0.4585 |
| 0.4551 | 27.7136 | 1500 | 0.4646 |
| 0.4516 | 29.5612 | 1600 | 0.4644 |
| 0.4471 | 31.4088 | 1700 | 0.4636 |
| 0.4427 | 33.2564 | 1800 | 0.4645 |
| 0.4494 | 35.1039 | 1900 | 0.4653 |
| 0.4357 | 36.9515 | 2000 | 0.4650 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.4.1+cu121
- Datasets 3.0.1
- Tokenizers 0.19.1