OpenAI is changing how we talk to ChatGPT. The company announced a fresh update for both the web and mobile app that moves voice conversations directly into your main chat feed. Previously, using Voice mode meant staring at a floating white circle on a blank screen. While that interface was clean, it isolated your spoken conversation from the rest of your history. Now, the AI responds with voice, text, and visuals all in one place.
To start talking, you click the small waveform icon next to the typing bar. Instead of taking over your whole screen, the AI keeps you in the current conversation window. This is a big help if you want to look back at what you just discussed or verify what the AI heard. As you speak, ChatGPT generates a live transcript of the answers, allowing you to read along while you listen.
The real benefit here is the addition of visual context. In a demo video released by the company, OpenAI showed how it works in practice. If you ask about the best bakeries in San Francisco, the AI won’t just list them out loud. It can pop up a map showing the locations and display photos of pastries from a specific shop like Tartine. You get the speed of a voice conversation with the rich detail of a text search.
This mix of audio and visual elements makes the assistant feel much more capable. Since you can already send images to the AI for analysis, it feels natural that it should send visuals back during a voice chat. Google is currently working on similar features with its Gemini AI to make conversations more expressive, suggesting this is the direction the industry is heading.
If you prefer the old way with the floating orb, you can still use it. OpenAI added a “Separate mode” switch in the settings for anyone who wants that specific interface. But for most users, this new layout offers a smoother experience. It removes the barrier between talking and typing. You can talk, watch the answers appear, review earlier messages, and look at maps without ever switching screens or losing your place in the chat.
