🔍 Read the full analysis: OpenAI Retires Two ChatGPT Voice Modes December 11: What Changes? on ThorstenMeyerAI.com
Get movie nights delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
TL;DR
OpenAI says it will retire Advanced and Standard voice from ChatGPT on Dec. 11, 2026, switching users who selected them to GPT-Live. Existing conversations will remain available, while live video and screen sharing will keep using Advanced voice until GPT-Live supports those features. Some users have complained about GPT-Live’s interruptions and voice differences.
OpenAI will retire ChatGPT’s Advanced and Standard voice modes on Dec. 11, 2026, leaving GPT-Live as the service’s general voice option, the company said in a Nov. 10 post on X. Users who selected either older mode will be switched to GPT-Live when the options are removed from Settings > Voice.
OpenAI’s help-center guidance, quoted by Interesting Engineering and FourWeekMBA, says users can continue using Advanced or Standard voice during the notice period. The company began notifying users on Nov. 10, giving them 31 days before the change. The modes will be removed from the voice settings on Dec. 11; OpenAI says users do not need to move their conversations manually.
Past conversations will remain in chat history and can be continued by text or voice. If users continue speaking, the new exchange will use GPT-Live. OpenAI has also listed exceptions: live video and screen sharing will continue to rely on Advanced voice until GPT-Live supports them, while voice with custom GPTs will use Advanced voice until those GPTs are deprecated on Dec. 11, according to help-center language reported by the outlets.
The switch is a change to ChatGPT’s consumer voice choices, not, based on the cited information, an announced shutdown of the voice API. OpenAI’s API deprecation page did not list a voice deprecation for Nov. 10. Separate older API voice and audio models do have announced shutdown dates from Jan. 6 to Feb. 26, 2027, according to the source material.
Two voice modes retire What changes on December 11?
OpenAI plans to remove Advanced and Standard voice from ChatGPT. GPT-Live becomes the general voice option for users who selected either mode.
What changes for users
Users on either retired mode will switch to GPT-Live when the options leave Settings > Voice.
Advanced and Standard voice remain available until the change date.
No manual move is needed; future spoken turns use GPT-Live.
Existing chats stay in history and can continue by text or voice.
How the modes handle speech
The retired modes and GPT-Live use different approaches to listening and responding.
| Mode | How it works |
|---|---|
| Standard | Speech recognition, language model, then speech generation |
| Advanced | Processes speech directly and waits for the user to stop |
| GPT-Live | Can listen and speak at the same time |
Exceptions and scope
- Live video and screen sharing keep using Advanced voice until GPT-Live supports them.
- Voice with custom GPTs uses Advanced voice until those GPTs are deprecated on Dec. 11.
- The announced retirement concerns voice modes in ChatGPT.
- The cited API deprecation page did not list a voice retirement for Nov. 10.
- Separate older voice and audio models have shutdown dates from Jan. 6 to Feb. 26, 2027.
What remains uncertain
No video timetable
OpenAI has not said when GPT-Live will support live video or screen sharing.
Reported complaints
Some users report interruptions, filler phrases, and voice differences; these are not published comparative test results.
Usage claims lack figures
OpenAI says voice use has “nearly doubled” and “virtually all” users now use GPT-Live, without figures in the cited material.
How the Voice Switch Affects Users
For people who use ChatGPT by speaking rather than typing, the change removes the ability to keep using a preferred older mode. That may matter to users who value a particular voice or interaction pattern, as well as people who rely on voice for accessibility. The source material reports complaints that GPT-Live interrupts, fills pauses with brief phrases such as “yeah” or “anyway,” and sounds different from the modes being retired.
OpenAI’s stated case for consolidation is usage: it said voice use has “nearly doubled” since GPT-Live launched in July and that “virtually all” ChatGPT voice users now use GPT-Live. Those are the company’s descriptions; it has not supplied figures in the cited material to show the user totals or explain how it measures the increase. The decision could simplify the voice options, but users who prefer the older modes will lose that choice even if they disagree with the company’s usage assessment.
The change also matters for organizations and developers, but its scope should not be overstated. The announced Dec. 11 retirement concerns voice modes in ChatGPT. The source material does not establish that this date ends API access to those modes. Developers should check the relevant API documentation for separate model retirements rather than assume the ChatGPT settings change applies to their products.
Popular right now
Bose QuietComfort True Wireless Earbuds with Active Noise Cancellation, Bluetooth 5.3, USB-C Charging, and IPX4 Water Resistance
Ninja Combi 14-in-1 Multicooker, Air Fryer, Toaster Oven & Steamer (SFP701)
iRobot Roomba Plus 405 (G181) 2-in-1 Vacuum and Mop Robot with AutoWash Dock
As an affiliate, we earn on qualifying purchases.
How GPT-Live Differs From Older Modes
The three modes use different approaches to speech. Standard voice linked speech recognition, a language model and speech generation in a sequence, according to help-center wording quoted by FourWeekMBA. Advanced voice processed speech directly but waited for a user to stop speaking before responding. GPT-Live is described in OpenAI’s developer documentation as a full-duplex voice model that can listen and speak at the same time.
OpenAI introduced GPT-Live on July 8 as a new generation of voice models for natural human-AI interaction. Its developer documentation says the API model can pass reasoning or tool use to a backend agent. Interesting Engineering reported that this allows GPT-Live to hand some work to a text model while continuing the conversation. OpenAI has said it plans to bring features from the older modes to GPT-Live and improve response depth, conversational continuity and personalization, according to the cited coverage.
The retirement overlaps with a separate plan to deprecate custom GPTs. The source material says OpenAI announced in September that it planned to retire custom GPTs across ChatGPT plans and move creators to plugins; an Enterprise migration FAQ gives Dec. 11 as the date custom GPTs stop running. The timing explains the stated exception for voice with custom GPTs, but the custom-GPT change is distinct from the removal of Advanced and Standard voice.
““I’m beyond disappointed because ChatGPT means a lot to me.””
— A user quoted in TechRadar’s Nov. 11 coverage
Open Questions About GPT-Live
OpenAI has not said when GPT-Live will support live video or screen sharing, so the length of the Advanced voice exception is unknown. The cited material also does not spell out whether the company will preserve every Advanced and Standard feature in GPT-Live or when the promised improvements to continuity and personalization will arrive.
There is also no firm measure of how many users oppose the change. TechRadar described the critics as a small but vocal group, while an OpenAI community-forum thread titled “Keep Standard Voice Mode!” had 100 posts when checked on Nov. 12, according to the source material. Those details show complaints but do not provide a representative survey. Reports of interruptions, filler phrases and voice differences are user feedback, not a published comparative test.
The help-center details in the source material were quoted by other outlets because the pages could not be machine-read when the source was prepared. OpenAI’s X post is the direct source for its announcement and explanation; some operational details rely on those reported help-center quotations.
What Users Should Expect by Dec. 11
Until Dec. 11, 2026, users can keep using Advanced or Standard voice, according to the reported help-center instructions. On that date, both options are scheduled to disappear from Settings > Voice, and users who had chosen them will move to GPT-Live. Their existing chat history is expected to remain available, with future spoken turns using the new mode.
Users who depend on live video or screen sharing should watch for an OpenAI update about GPT-Live support, since no timetable has been provided. Developers should separately review OpenAI’s API deprecation notices: the ChatGPT voice retirement is not, on the evidence cited here, an API shutdown announcement, while other older audio models have dates listed for early 2027.
Key Questions
Which ChatGPT voice modes are being retired?
Advanced voice and Standard voice are scheduled for removal from ChatGPT on Dec. 11, 2026. GPT-Live will remain as the general voice option.
Will my old voice conversations be deleted?
No deletion is described in the reported help-center guidance. Existing conversations will remain in chat history and can be continued by text or voice; spoken continuations will use GPT-Live.
Will GPT-Live support live video and screen sharing on Dec. 11?
That is not confirmed. OpenAI’s reported guidance says those features will continue using Advanced voice until GPT-Live has the capability, but gives no release date.
Does this retire OpenAI’s voice API?
The announced change applies to voice modes in ChatGPT. The cited API deprecation page did not list a voice retirement for Nov. 10, though separate older voice and audio models have shutdown dates in January and February 2027.
Source: ThorstenMeyerAI.com
