https://www.youtube.com/watch?v=Vi-jybt-KTg
This fork utilizes OpenAI's Realtime API in combination with the Simli Avatar API to enable avatar interaction. While still in development, it essentially facilitates video agent conversations.
This project demonstrates how to integrate OpenAI's Realtime Voice-to-Voice API with Simli's avatar technology to create interactive video conversations with an AI agent represented by a visual avatar.
get_weather()
with map display.set_memory()
to store and retrieve user-specific data.Clone the Repository
git clone https://github.com/yourusername/yourproject.git
cd yourproject
npm install
Create a .env File
In the root directory of the project, create a file named .env. Add Your API Keys Add your OpenAI and Simli API keys to the .env file:
REACT_APP_OPENAI_API_KEY=your_openai_api_key
REACT_APP_SIMLI_API_KEY=your_simli_api_key
REACT_APP_VOICE=your_openai_api_Voice
REACT_APP_SIMLI_FACE_ID=your_simli_face_id
Replace your_openai_api_key and your_simli_api_key with your actual API keys. Note: Keep your API keys secure and do not commit the .env file to version control.
Start the Application npm run build npm run start The application will run on http://localhost:3000. Interact with the Assistant
Connect: Click on the Connect button to establish a connection. Push to Talk: Use the Push to Talk (Hold while speaking) button to start speaking to the assistant. View Avatar: The Simli avatar will appear, representing the AI assistant. Use Tools: Try out tools like get_weather() and set_memory() within the conversation. "# Simli_GPT4o-realtime"