TeachMe

What you'll need to run the frontend:

• Bun

What you'll need to run the server:

• Python >3.10
• Pip
• Uvicorn
• Ngrok
• OpenAI API Key
• Retell API Key and a Retell Agent

Run the frontend

1. Go to the frontend directory

cd frontend

2. Install the dependencies

bun install

3. Start the frontend

bun dev

4. Open in your browser: http://localhost:3000

Run the server

1. Navigate to the backend directory

cd backend

2. Install the requirements

pip install -r requirements.txt

3. Add your OpenAI API and Retell API keys to the .env file

OPENAI_API_KEY=[YOUR_OPEN_AI_KEY]
RETELL_API_KEY=[YOUR_RETELL_AI_KEY]

4. Start server

uvicorn main:app --reload

5. Tunnel the server to the internet using ngrok

ngrok http 8000

6. Set the Custom LLM URL for your Retell agent the ngrok url + /voice/llm-websocket

Example:

https://e253-128-195-97-137.ngrok-free.app/voice/llm-websocket