Yc1981/yochat-ai

AI-powered English speaking platform for Tunisian middle school students. # YoChat Deploy-Ready Version This cleaned version separates the project into two deployable parts: - `frontend/` — React + Vite user interface - `backend/` — FastAPI WebSocket backend for Gemini Live audio conversation The old Express/Node backend files were removed from the active app structure to avoid deployment conflicts. ## 1. Local backend test ```bash cd backend python -m venv .venv # Windows: .venv\Scripts\activate source .venv/bin/activate pip install -r requirements.txt cp .env.example .env # add your GEMINI_API_KEY in .env uvicorn main:app --host 0.0.0.0 --port 8000 --reload ``` Open: ```text localhost ``` Expected result: ```json {"status":"healthy","key_configured":true} ``` ## 2. Local frontend test ```bash cd frontend npm install cp .env.example .env.local npm run dev ``` Open: ```text localhost ``` ## 3. Deploy backend on Render Create a new **Web Service** from GitHub. Use these settings: ```text Root Directory: backend Build Command: pip install -r requirements.txt Start Command: uvicorn main:app --host 0.0.0.0 --port $PORT ``` Add this environment variable in Render: ```text GEMINI_API_KEY=your_real_key ``` After deploy, test: ```text your-render-service.onrende… ``` ## 4. Deploy frontend on Vercel Create a new Vercel project. Use these settings: ```text Root Directory: frontend Build Command: npm run build Output Directory: dist ``` Add this environment variable in Vercel: ```text VITE_BACKEND_URL=your-render-service.onrende… ``` Do not add `/ws/live` at the end. The React app adds it automatically. ## 5. Important deployment note For microphone and WebSocket audio to work online, the frontend must be on HTTPS and the backend WebSocket must use WSS. This is handled automatically when `VITE_BACKEND_URL` starts with `https://`.