metadata
license: mit
base_model: MBZUAI/speecht5_tts_clartts_ar
tags:
- generated_from_trainer
datasets:
- common_voice_11_0
model-index:
- name: ZiadTTS
results: []
ZiadTTS
This model is a fine-tuned version of MBZUAI/speecht5_tts_clartts_ar on the common_voice_11_0 dataset. It achieves the following results on the evaluation set:
- Loss: 0.6477
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: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- 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: 500
- training_steps: 4000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.9355 | 1.45 | 1000 | 0.8428 |
| 0.7083 | 2.9 | 2000 | 0.6772 |
| 0.6844 | 4.36 | 3000 | 0.6527 |
| 0.6732 | 5.81 | 4000 | 0.6477 |
Framework versions
- Transformers 4.38.0.dev0
- Pytorch 2.1.0+cu121
- Datasets 2.17.1
- Tokenizers 0.15.2