8934528/Com

voice communication with memory, tts, stt, and translation for south african languages. # Voice Communication Assistant An AI-powered voice communication system with persistent memory, featuring Speech-to-Text and Translation capabilities for South African languages. --- ## Features - **Voice Recognition** - Convert speech to text using speech recognition STT API - **Multi-language Support** - English, isiZulu, Sesotho, Afrikaans, isiXhosa, and more - **Real-time Translation** - Translate between South African languages - **Persistent Memory** - Conversations saved to JSON - **Text-to-Speech** - Multiple voice options (male, female, robotic) - **Modern UI** - Bootstrap 5 with Flaticon icons - **WebSocket Support** - Real-time bidirectional communication ## Supported Languages ### Speech-to-Text (STT) - South African English (`eng`) - isiZulu (`zul`) - Sesotho (`sot`) - Afrikaans (`afr`) - African French (`fra`) - Code-switched isiZulu (`cs-zul`) ### Translation - English (`eng_Latn`) - isiZulu (`zul_Latn`) - isiXhosa (`xho_Latn`) - Afrikaans (`afr_Latn`) - Sesotho (`sot_Latn`) - Kiswahili (`swh_Latn`) --- ## Installation ### Prerequisites - Python 3.8+ - pip - PortAudio (Linux/Mac) or appropriate audio drivers (Windows) ### Quick Start 1. **Clone the repository** bash git clone cd Com 2. Set up environment variables bash cp .env.example .env # Edit .env and add your Vulavula API key 3. Run the application `**Linux/Mac:** bash chmod +x run.sh ./run.sh `**Windows:** bash run.bat 4. Open your browser Navigate to `localhost` --- ## Project Structure Com/ ├── static/ │ ├── style.css # Custom styling │ ├── script.js # Client-side JavaScript │ ├── settings.js # Settings panel JavaScript │ └── audio/ # Audio storage directory ├── templates/ │ ├── index.html # Main Frontend UI │ └── settings.html # Settings Modal Template ├── com.py # Main Flask application ├── requirements.txt # Python dependencie …