Darijatts V0.1 500M

🚀 Moroccan Darija TTS

This is a text-to-speech (TTS) model for Moroccan Darija, offering high - quality voice synthesis for this language.

🚀 Quick Start

This is a text-to-speech (TTS) model for Moroccan Darija, fine-tuned from OuteAI/OuteTTS-0.2-500M on the KandirResearch/DarijaTTS-clean dataset.

✨ Features

• Model Adaptation: Fine - tuned from a pre - trained model to suit Moroccan Darija.
• Dataset - Based Training: Trained on a clean and relevant dataset for better performance.
• Efficient Training: Utilized LoRA - based fine - tuning for efficiency.

📦 Installation

You need to install outetts and llama - cpp - python to run the model.

pip install outetts llama-cpp-python huggingface_hub

💻 Usage Examples

Basic Usage

import outetts
from outetts.models.config import GenerationConfig
from huggingface_hub import hf_hub_download

model_path = hf_hub_download(
    repo_id="KandirResearch/DarijaTTS-v0.1-500M",
    filename="unsloth.Q8_0.gguf",
)
model_config = outetts.GGUFModelConfig_v2(
    model_path=model_path,
    tokenizer_path="KandirResearch/DarijaTTS-v0.1-500M",
)
interface = outetts.InterfaceGGUF(model_version="0.3", cfg=model_config)

def tts(text, temperature=0.3, repetition_penalty=1.1):
    gen_cfg = GenerationConfig(
        text=text,
        temperature=temperature,
        repetition_penalty=repetition_penalty,
        max_length=4096,
    )
    output = interface.generate(config=gen_cfg)
    output_path = "output.wav"
    output.save(output_path)
    return output_path

# Example usage
audio_path = tts("السلام كيداير لاباس عليك؟")
print(f"Generated audio saved at: {audio_path}")

📚 Documentation

Model Details

Training

The model was fine - tuned using Unsloth's SFTTrainer. The dataset was preprocessed following the OuteTTS training guide. LoRA - based fine - tuning was applied to improve efficiency.

📄 License

This project is licensed under the apache - 2.0 license.

Support Me

For any issues or improvements, feel free to open a discussion or PR!