OpenAI is upgrading ChatGPT’s voice mode with a new model called GPT-Live-1, designed to make spoken conversations feel more natural and less interruptive. According to reporting from multiple outlets, OpenAI positions the update as closer to “talking to another person,” including behavior changes during back-and-forth speaking. OpenAI says GPT-Live-1 interrupts users less, and when a user pauses mid-conversation, it waits for the user to continue rather than taking over immediately.
The rollout reaches users worldwide, with outlets reporting that GPT-Live-1 and a smaller “GPT-Live-1 mini” are available starting on 8 July. In addition to improved turn-taking, one outlet reports the voice update supports more natural conversation features such as simultaneous translation. OpenAI presented details during a press briefing, where an OpenAI researcher described GPT-Live-1 as the company’s upgraded voice model.
Overall, the sources agree this update is meant to improve the flow of voice interactions—reducing interruptions, responding to pauses appropriately, and expanding capabilities like translation.
