Generate Next.js pages using Voice (OpenAI Realtime API)

Fork of the OpenAI Realtime Console with modifications to edit a live codesandbox voice-interactive using API reference for the OpenAI Realtime API. It comes packaged with two utility libraries, openai/openai-realtime-api-beta that acts as a Reference Client (for browser and Node.js) and /src/lib/wavtools which allows for simple audio management in the browser.

Recording.2024-10-08.111409.mp4

https://www.loom.com/share/6635cb9654da4baeb36a1ef67e5fe085?sid=3998f3c4-4b35-4c45-ae01-016acc13e422

Set your own API Key

1. Set your API key (saved to local storage)
2. Click "Connect" and start talking
3. Watch LLM generated code appear and preview in a Nodebox

This is a React project created using create-react-app that is bundled via Webpack. Install it by extracting the contents of this package and using;

$ npm i

Start your server with:

$ npm start

It should be available via localhost:3000.

Acknowledgements and contact

Creators of the original console project:

• OpenAI Developers / @OpenAIDevs
• Jordan Sitkin / API / @dustmason
• Mark Hudnall / API / @landakram
• Peter Bakkum / API / @pbbakkum
• Atty Eleti / API / @athyuttamre
• Jason Clark / API / @onebitToo
• Karolis Kosas / Design / @karoliskosas
• Keith Horwood / API + DX / @keithwhor
• Romain Huet / DX / @romainhuet
• Katia Gil Guzman / DX / @kagigz
• Ilan Bigio / DX / @ilanbigio
• Kevin Whinnery / DX / @kevinwhinnery