If you talk to ChatGPT out loud, the voice on the other end is about to change for good. OpenAI says it will retire ChatGPT’s two older voice modes, called Standard Voice and Advanced Voice, on December 11, 2026. After that date, everyone who uses voice in ChatGPT will be moved to GPT-Live, the newer voice system the company introduced in July.
OpenAI began notifying users of the change on November 10, according to Interesting Engineering, and the news has spread quickly across tech sites and social media this week. Some longtime users are not happy about it.
What is changing on December 11
Until December 11, people who still have Standard or Advanced Voice selected can keep using them. After that, ChatGPT will switch them to the Live voice automatically, and the older options will no longer appear in settings, according to reports from Interesting Engineering, Android Headlines and Flickonclick. Users do not need to move anything themselves. Past voice conversations stay in chat history and can be continued by text or by voice.
Analytics Insight reports the change applies across the web and the ChatGPT apps for iPhone and Android.
There is one important exception. Live video and screen-sharing will keep running on Advanced Voice for now, because GPT-Live cannot yet handle visual input. OpenAI has not given a timeline for when the newer system will support those features, according to Interesting Engineering and Analytics Insight.
Reports differ slightly on custom GPTs, the personalized chatbots people build inside ChatGPT. Interesting Engineering and Android Headlines say they lose the older voice options on December 11, so anyone who built one around a particular voice may need to update it. Readers who depend on custom GPTs should check OpenAI’s help center for the latest details.
How GPT-Live is different
The three systems work in different ways. Standard Voice chained three steps together: it turned your speech into text, wrote a reply, then read that reply aloud. Advanced Voice processed audio directly but still waited for you to finish talking before it answered.
GPT-Live is designed to feel more like a real conversation. According to the reports, it can listen and speak at the same time, lets you interrupt mid-sentence, treats short sounds like “yeah” or “mhmm” as signs you are listening rather than interrupting, and tries not to rush in when you pause to think. It can also hand harder, multi-step questions to OpenAI’s stronger text models in the background without ending the voice chat. Analytics Insight reports it can use web search, memory and connected apps inside the same conversation.
OpenAI says nearly all voice users have already moved to GPT-Live and that voice use has nearly doubled since it launched, according to Interesting Engineering. The company says running one voice system instead of several will let it fix bugs and add features faster.
Why some users are frustrated
Not everyone wants the upgrade. Android Headlines reports frustration on Reddit and X from longtime users who say GPT-Live answers too quickly and cuts off their thoughts, and some say the voices that carry over feel unfamiliar. The site notes that OpenAI acknowledged in its help center that switching voice styles can feel jarring.
OpenAI has backed off retirement plans after user pushback before, Android Headlines points out, so the December 11 date could still change. Flickonclick reports OpenAI says feedback from users of the older modes will shape future Live voice updates, including more detailed answers and longer conversations.
The move fits a wider push by tech companies to make AI assistants feel more natural to talk to. Google recently opened its most powerful AI model, Gemini 4 Carbon, to paying users, and Apple has been opening Siri’s AI to app developers.
Why it matters for families
- No action is required. If you use ChatGPT voice, you will be switched automatically on December 11. Your old conversations are not deleted.
- Expect a different feel. The new voice responds faster and may sound different. If you or an older relative relies on voice for help, try Live voice now so the change is not a surprise.
- Video and screen-sharing still work on the older system for the time being.
- Kids and teens using voice chat should know that a faster, more natural-sounding AI is still a computer program. Parents may want to review how their children use AI assistants and what they share.
Upstormed News will keep following the rollout and any changes to the December 11 date.
Read more: Anthropic Sets IPO Price Range of $70-$78, Eyeing Up to $1.4 Trillion Valuation
Do you talk to AI assistants out loud, or do you prefer typing? Tell us in the comments.
Sources: Interesting Engineering, “OpenAI pulls the plug on two major voice modes as GPT-Live takes over” (November 10, 2026); Android Headlines, “ChatGPT Is Killing Off Standard Voice Modes and Users Are Frustrated” (November 11, 2026); Analytics Insight, “ChatGPT Voice Overhaul: OpenAI to Replace Advanced Modes with Live” (November 12, 2026); Flickonclick, “ChatGPT Advanced Voice Ending On December 11: What Changes” (November 12, 2026); OpenAI Help Center, “Retiring Advanced and Standard Voice in ChatGPT”
Leave a Reply