OpenAI Launches GPT-Live Voice That Listens and Speaks at Once
OpenAI has released GPT-Live, a pair of voice models that can listen and talk at the same time, replacing the older Advanced Voice Mode in ChatGPT.
A more natural voice
OpenAI has launched GPT-Live, a pair of new voice models that change how people talk to ChatGPT. Unlike the older system, the new models can listen and speak at the same time, closer to how a real conversation flows.
Full duplex, not walkie-talkie
The previous voice feature worked in turns, with the app either listening or talking. GPT-Live uses a full-duplex design, so it can pick up interruptions, react mid-sentence and keep pace with a back and forth exchange.
Replacing Advanced Voice Mode
The models replace the existing Advanced Voice Mode. OpenAI describes the shift as a fundamental redesign rather than a small upgrade, aimed at making spoken chats feel less robotic.
Where it helps
More natural voice could make ChatGPT more useful for hands-free tasks, accessibility, language practice and customer support, where stiff turn-taking has long been a weak point.
