Logo Lanfrica

youngstunners88/SA-voices

Domaine:

natural language processing

Type de record:

software
Créateur:
you
HĂ´te:
🇿🇦 South African Multilingual Voice Agent - Supporting all 11 official languages with Qwen3-TTS, error correction, and self-improvement # 🇿🇦 SA Voices - South African Multilingual Voice Agent > **A state-of-the-art voice agent supporting all 11 official South African languages using Qwen3-TTS** ## 🎯 Mission SA Voices democratizes voice technology for South Africa by providing high-quality, culturally-aware text-to-speech in: - **isiZulu** (Zulu) - **isiXhosa** (Xhosa) - **Afrikaans** (Afrikaans) - **English** (English) - **Sepedi** (Northern Sotho) - **Sesotho** (Southern Sotho) - **Setswana** (Tswana) - **Xitsonga** (Tsonga) - **siSwati** (Swati) - **Tshivenda** (Venda) - **isiNdebele** (Ndebele) ## 🏗️ Architecture ``` sa-voices/ ├── src/ │ ├── core/ # Core voice agent engine │ ├── routing/ # Intelligent request routing │ ├── state/ # State management system │ ├── tts/ # Qwen3-TTS integration │ ├── dataset/ # WaxalNLP dataset handlers │ ├── ui/ # Web interface │ ├── api/ # REST/WebSocket API │ └── utils/ # Utilities ├── skills/ # Specialized agent skills ├── tests/ # Test suites ├── docs/ # Documentation ├── config/ # Configuration files ├── scripts/ # Setup and utility scripts └── examples/ # Usage examples ``` ## 🚀 Quick Start ### Prerequisites ```bash # Install system dependencies sudo apt-get update sudo apt-get install ffmpeg # Python 3.10+ required python --version ``` ### Installation ```bash # Clone the repository git clone github.com cd SA-voices # Create virtual environment python -m venv venv source venv/bin/activate # On Windows: venv\Scripts\activate # Install dependencies pip install -r requirements.txt # Install Qwen3-TTS git clone github.com cd Qwen3-TTS pip install -e . cd .. # Install audio dataset support pip install datasets[audio] ``` ### Environment Setup ```bash # Copy environment template cp .env.example .env # Edit .env with your c …