How we built a realtime system for responsive voice AI in six months

OpenAI said it built a realtime system for responsive voice AI in six months. The system, called GPT-Live, is the company's third-generation voice system and removes the turn detector from the audio path. Its voice model is full-duplex, meaning it can listen and speak at the same time. That eliminates the need for a separate detector and makes conversation feel more immediate and natural. When deeper reasoning or tool use is needed, GPT-Live can also consult frontier models such as GPT-5.5 without interrupting the flow of the conversation. Delivering this experience at scale required a new system architecture optimized for low latency. The system streams incoming audio into the voice model and outbound speech back to the user while handling delegation on a separate asynchronous path. Over the last six months, engineers reworked model inference, context management, and media transport to keep speech flowing smoothly from end to end. The architecture also creates a clean boundary between the core voice path and application logic. This foundation powers a growing range of capabilities in ChatGPT Voice, including the newly launched ability to control your computer and coordinate your agents in the ChatGPT desktop app. The team wrote the media frontend and inference logic in Go, replacing a previous Python asyncio implementation. This significantly improved the smoothness of frame delivery, with the new system's p95 matching the previous system's p50.