SIAgentNews.com
← All news
DevelopingOpenAI

OpenAI will retire ChatGPT's Advanced and Standard voice modes on December 11

ChatGPT's two older voice options are going away, leaving GPT-Live as the only choice. Here is what changes, what stays, and what to check before the date.

· First reported by Interesting Engineering

December 11, 2026Date Advanced and Standard voice disappear from ChatGPT

The short version

  • Advanced and Standard voice leave ChatGPT on December 11, 2026; GPT-Live becomes the default.
  • OpenAI says nearly all voice users have already moved to GPT-Live, and voice usage has nearly doubled since July.
  • Custom GPTs lose Advanced voice on the same date.
  • Live video and screen-sharing still run on Advanced voice, with no timeline to move them.
  • Past chats stay in your history and can continue by text or Live voice.

Why it matters

If you talk to ChatGPT instead of typing, the voice you are used to is about to be replaced, so it is worth trying the new one while the old ones still work. If you use voice with custom GPTs, or rely on video and screen-sharing, the date affects you in different ways.

What OpenAI is removing

OpenAI is shutting down two older ways of talking to ChatGPT. On December 11, 2026, Advanced voice and Standard voice will disappear from the voice settings, and GPT-Live will be the default voice experience. As Interesting Engineering reported, OpenAI began notifying users on November 10.

Until the date arrives, people who prefer the older modes can keep using them. After it passes, ChatGPT will switch affected users to GPT-Live automatically. Nobody has to move their old conversations by hand or start new chats.

OpenAI says the change should not come as a shock to most people. By its account, nearly all ChatGPT voice users have already moved to GPT-Live since it launched in July. Voice usage has also nearly doubled over that stretch.

Three generations of talking to a chatbot

The three voice systems work in quite different ways, and the differences explain why OpenAI wants to drop the old ones.

  • Standard voice chained together three separate models. One turned your speech into text, a second wrote the reply, and a third turned that reply into audio. Each handoff added delay, and some information got lost between audio and text.
  • Advanced voice used one model that handled audio going in and coming out directly. It still had to wait until you finished speaking before it could answer.
  • GPT-Live is built differently. It can listen and speak at the same time, so conversation does not have to go strictly one turn after another.

In practice, that means quicker exchanges and interruptions that feel more natural. GPT-Live can also give a short acknowledgment while you are still talking. It can stay quiet if you need a moment to think, which makes the conversation feel less rigid.

Handing off the hard questions

One more feature stands out. GPT-Live connects to OpenAI's text-based models for harder requests. When you ask something complicated or a task with several steps, it can pass the work along without ending the voice conversation.

You can keep talking while the heavier work happens in the background. That is a different idea from a voice mode that only reads out what a text chatbot wrote. OpenAI says it also plans to improve response depth, conversational continuity and personalization as GPT-Live develops.

What changes for you, and what does not

For everyday users, the practical effects are small but real. Advanced and Standard voice will vanish from the Voice section of Settings after December 11. Past conversations will stay in your ChatGPT history, and you can pick them up again through text or through the new Live voice experience.

Custom GPTs, the tailored versions of ChatGPT that people build and share, are affected too. They will lose access to Advanced voice on December 11. Until then, conversations with them keep using the older model. The source advises that developers and users should plan for this if they have voice workflows built around those GPTs.

The shift is not total, though. Live video and screen-sharing will keep running on Advanced voice, because GPT-Live does not support them yet, according to OpenAI. The company says it will maintain the older system until the newer one can do the same. It has not given a date for that.

So December 11 ends the two voice modes you can pick, not every feature that Advanced voice powers.

What to do before the deadline

The simplest move is to try GPT-Live on a task you really do, such as dictating an email, talking through an idea, or practicing a conversation. That way you find out whether the interruptions and pauses suit you while the old modes are still available as a fallback.

If you build or use custom GPTs with voice, check how they behave. If you depend on video or screen-sharing in ChatGPT, note that those stay on the older system for now, but nothing says for how long.

What to watch

  • When will GPT-Live support live video and screen-sharing? OpenAI has given no timeline.
  • How will custom GPTs that rely on voice behave once Advanced voice is removed?
  • Will the planned improvements to depth, continuity and personalization arrive in a way users notice?

Try it

Open ChatGPT's voice feature this week and use GPT-Live for a real task, such as dictating an email, before the older modes disappear on December 11.

Sources and links

First reported by Interesting Engineering. Written by the SI Agent News Desk from the reporting above.Written with AI from the source above. Every number is checked against the source. Spot a mistake? Corrections.

Liked it? Say why

Add a line for other readers.

No account needed. We never show your email. Comments with links, ads or abuse are held and not shown. Your comment shows after a quick check.

More AI news