---
library_name: transformers
language:
- ar
license: cc-by-nc-4.0
base_model: MBZUAI/speecht5_tts_clartts_ar
tags:
- generated_from_trainer
datasets:
- bashiralsuty/sudanese_dialect_sample
model-index:
- name: Speecht5 TTS Sudanese Arabic
  results: []
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# Speecht5 TTS Sudanese Arabic

This model is a fine-tuned version of [MBZUAI/speecht5_tts_clartts_ar](https://huggingface.co/MBZUAI/speecht5_tts_clartts_ar) on the sudanese_dialect_sample dataset.
It achieves the following results on the evaluation set:
- Loss: 1.0462

## 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: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- 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 |
|:-------------:|:---------:|:----:|:---------------:|
| 0.5169        | 333.3333  | 1000 | 0.8784          |
| 0.4347        | 666.6667  | 2000 | 0.9882          |
| 0.3989        | 1000.0    | 3000 | 1.0438          |
| 0.3888        | 1333.3333 | 4000 | 1.0462          |


### Framework versions

- Transformers 4.48.0
- Pytorch 2.8.0+cu126
- Datasets 3.6.0
- Tokenizers 0.21.4