metadata
language:
- sd
license: mit
base_model: microsoft/speecht5_tts
tags:
- generated_from_trainer
datasets:
- facebook/voxpopuli
model-index:
- name: SpeechT5 TTS sindhi
results: []
SpeechT5 TTS sindhi
This model is a fine-tuned version of microsoft/speecht5_tts on the google/fleurs dataset. It achieves the following results on the evaluation set:
- Loss: 0.3386
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: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 50
- training_steps: 500
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.3716 | 0.2703 | 50 | 0.3473 |
| 0.3764 | 0.5405 | 100 | 0.3442 |
| 0.3736 | 0.8108 | 150 | 0.3438 |
| 0.3707 | 1.0811 | 200 | 0.3438 |
| 0.3736 | 1.3514 | 250 | 0.3427 |
| 0.3717 | 1.6216 | 300 | 0.3414 |
| 0.3614 | 1.8919 | 350 | 0.3401 |
| 0.3641 | 2.1622 | 400 | 0.3393 |
| 0.3679 | 2.4324 | 450 | 0.3394 |
| 0.3644 | 2.7027 | 500 | 0.3386 |
Framework versions
- Transformers 4.45.0.dev0
- Pytorch 2.3.1+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1