This project provides a Next.js 15 starter template for building real-time voice AI applications using OpenAI's Realtime API and WebRTC. It targets developers looking to create interactive voice experiences with features like tool-calling and localization, leveraging modern UI components from shadcn/ui and Framer Motion.

How It Works

The application utilizes a WebRTC-based architecture for real-time audio streaming between the client and OpenAI's API. A custom hook abstracts the complexities of WebRTC handling, simplifying integration. It supports OpenAI's tool-calling capabilities with six example functions and offers localization for multiple languages, enhancing user experience and accessibility.

Quick Start & Requirements

• Install Dependencies: npm install (Node.js) or deno install (Deno).
• Run: npm run dev (Node.js) or deno task start (Deno).
• Requirements: Node.js or Deno runtime, OpenAI API Key or Azure OpenAI API Key (configured in .env).
• Demo: https://github.com/user-attachments/assets/ea9324af-5c18-48d2-b980-2b81baeea4c0

Highlighted Details

• Next.js 15 framework with API routes.
• UI powered by shadcn/ui, Tailwind CSS, and Framer Motion.
• WebRTC handling abstracted via a custom hook.
• OpenAI tool-calling with 6 example functions.
• Localization support for English, Spanish, French, and Chinese.

Maintenance & Community

The project is a starter template, with the core WebRTC and Next.js implementation originating from skrivov/openai-voice-webrtc-next. No specific community channels or active maintenance signals are detailed in the README.

Licensing & Compatibility

Licensed under the MIT License, permitting commercial use and integration with closed-source projects.

Limitations & Caveats

The scrapeWebsite tool-calling example requires a FireCrawl API key. The project is presented as a starter template, implying potential for further development and refinement.