OpenAI to retire Standard and Advanced voice modes on December 11

What OpenAI is changing

OpenAI will retire the Standard and Advanced voice modes in ChatGPT on December 11, according to TechRadar. After that date, Live Voice will be the only voice option, and the change applies across all of OpenAI's subscription tiers.

Voice modes are not about dictation. They let a person hold a spoken, real-time conversation with the chatbot: the user talks, and ChatGPT answers aloud. Until the switch, users can pick among the three modes by opening Settings, then Voice.

How the three modes differ

Standard mode turns speech into text, generates a written reply and then reads it back. Advanced mode processes speech directly but alternates between listening and speaking, so the user cannot talk over it.

Live mode is the newest of the three. It can listen and speak at the same time, which allows more natural interruptions and quicker back-and-forth. OpenAI says nearly all ChatGPT users have already moved to Live, which is the company's stated reason for consolidating on it.

Why some users are upset

A smaller but vocal group has complained on Reddit and X, TechRadar reports. One user wrote that they were beyond disappointed because ChatGPT means a lot to them, while another described the situation as paying a monthly subscription to beta-test questionable decisions.

The objections centre on two things. Some people have grown attached to particular voices; one Reddit poster said they had used the Vale voice for years and found it comforting, and that choosing Vale in Live does not reproduce the same experience. Others find Live less natural than advertised, because it fills pauses with interjections, interrupts, and shifts tone more often, which some users find irritating.

OpenAI's response and a past reversal

In a statement on its Help Center, OpenAI acknowledged that some users are used to the style and responses of the older modes and said such model transitions can be difficult. The company pointed to its Help Center guidance on navigating model transitions.

This is not the first attempt to remove Standard Voice. In September 2025, OpenAI said it would retire the mode within 30 days, then reversed course on September 9, saying users valued the older experience and that it would stay available while the company addressed feedback. The current announcement appears more final, though TechRadar notes that continued resistance could test that.

What it means for users

The practical effect is that anyone who relies on Standard or Advanced will have to switch to Live after December 11, and users who picked a specific voice should expect a different experience even if a voice with the same name exists in Live. TechRadar frames the move as OpenAI concentrating development effort on one voice system instead of maintaining three.

Voice interaction itself is not going away. The question is whether the complaints grow loud enough to prompt another rethink, as happened in September 2025.