A web application that allows users to interact with OpenAI's latest GPT Realtime model using WebRTC for speech-to-speech conversations.
- Realtime speech-to-speech conversations with OpenAI's GPT Realtime model
- Support for OpenAI's exclusive voices (Cedar and Marin)
- Dark/light theme toggle
- Conversation history display
- Connection status indicators
- Persistent user preferences
- Low-latency audio streaming with WebRTC
- Direct audio input/output support
- An OpenAI API key with access to the GPT Realtime model
- A modern web browser that supports WebRTC
- Visit the live demo or host the application yourself
- Enter your OpenAI API key
- Choose a voice for the AI responses (Cedar or Marin)
- Click "Start Conversation" to begin
- Speak naturally to interact with the AI in realtime
- Click "Stop" when you're finished
This application runs entirely in your browser. Your OpenAI API key and conversations are not stored on any server and are only used to communicate with OpenAI's API.
This project is licensed under the MIT License - see the LICENSE file for details.
- OpenAI for providing the realtime voice API
- WebRTC for enabling browser-based audio streaming