ChatGPT Retires Advanced, Standard Voice Modes

// 24m agoPRODUCT UPDATE

ChatGPT Retires Advanced, Standard Voice Modes

OpenAI will retire ChatGPT’s Advanced and Standard voice modes on December 11, 2026, moving users to GPT-Live by default. The company says voice usage has nearly doubled since GPT-Live launched, with virtually all voice users now using it. [OpenAI Help Center](https://help.openai.com/en/articles/20001638-retiring-advanced-and-standard-voice-in-chatgpt)

// ANALYSIS

This is a pragmatic consolidation, but not a clean upgrade for everyone: OpenAI is removing user choice while GPT-Live still lacks some legacy capabilities.

–GPT-Live’s full-duplex design supports interruptions, natural pauses, and faster back-and-forth conversation.
–Existing Advanced and Standard conversations remain available in history and can continue through text or voice.
–Video and screen sharing will continue relying on Advanced voice until GPT-Live supports them.
–Developers and power users should retest workflows because GPT-Live’s conversational behavior differs from Standard’s predictable speech-to-text, LLM, and text-to-speech pipeline.
–The move reinforces OpenAI’s strategy of making voice a unified ChatGPT interface rather than a collection of separate modes. [OpenAI GPT-Live announcement](https://openai.com/index/introducing-gpt-live/)

// TAGS

gpt-livevoice-agentspeechmultimodalchatbot

DISCOVERED

24m ago

2026-11-11

PUBLISHED

1h ago

2026-11-11

RELEVANCE

8/ 10

AUTHOR

mark_k

// KEEP READING

More AI developer news from the feed

EXPLORE FULL FEED

UPDATE1h ago

ElevenLabs Plans Enterprise Model Fine-Tuning

ElevenLabs plans to let enterprises fine-tune selected text-to-speech, speech-to-text, and open-source language models using their own deployment data. Customization will target business-specific outcomes such as customer satisfaction and reliability.

MODEL1h ago

Google Drops Gemini Decision 1

Google released Gemini Decision 1 and its Decisions API for calibrated binary, categorical, and rubric decisions across text, images, audio, video, and PDFs, with zero output tokens billed. It targets routing, policy checks, ranking, and evaluation rather than conversational generation.

INFRA1h ago

Mastra Platform launches $25 Starter plan

Mastra’s new $25/month Starter plan includes an always-on Small dedicated server, two environments, usage alerts, 250K observability events, 10 GB egress, 10 connections, and 60-day trace retention. It positions dedicated infrastructure as the paid upgrade from Mastra’s free serverless tier. [Official announcement](https://mastra.ai/blog/mastra-platform-starter-plan)