ChatGPT Live Dictation Shows Your Words as You Speak and Drops the 10-Minute Limit

OpenAI rolled out Live Dictation in ChatGPT, which streams your speech into text in real time and removes the 10-minute recording limit. It is rolling out globally to all ChatGPT plans on iOS, Android, web and desktop.

Christopher Ort
By Christopher Ort · 2 min read
Share
A person dictating into a smartphone while walking down a city street.
Image: AI-generated, i10X
On this page
  1. Quick Take
  2. In plain English
  3. What happened
  4. How it works
  5. Why it matters
  6. What's still open
  7. Sources
  8. i10x Perspective

ChatGPT’s dictation now types while you talk. OpenAI rolled out Live Dictation on 18 November 2026, a faster speech-to-text mode that shows your words as text in real time, and it removed the old 10-minute recording limit at the same time. The update is rolling out globally to all ChatGPT plans, including free, on iOS, Android, web and the desktop app.

i10X Quick Take

Quick Take

  1. What’s new: speech turns into text as you speak, instead of after you stop recording.
  2. No time cap: the 10-minute recording limit is gone.
  3. Control: tap or click the waveform to show or hide the text while you talk.
  4. Who gets it: all ChatGPT plans, rolling out globally on iOS, Android, web and desktop.
  5. Not Voice mode: Dictation still produces text you can edit before sending. Voice is the spoken back-and-forth conversation.

In plain English

Before, you pressed the microphone, spoke, and then waited while ChatGPT turned your recording into text. Now the text appears as you go, so you can see right away if something was misheard. Long voice notes, meeting thoughts or detailed prompts no longer get cut off after ten minutes.

What happened

OpenAI announced the change in its ChatGPT release notes: “Dictation in ChatGPT is getting faster and more reliable. With Live Dictation, your speech turns into text as you talk, so you spend less time waiting for your message to finish transcribing.” The notes add: “We’ve also removed the 10-minute recording limit.”

Atty Eleti, who works on voice features at OpenAI, summarized the update on X: “Dictation now streams text live as you speak. Faster submission, improved reliablity, and no more 10m limit.”

How it works

  • Press the microphone icon in the message box and start speaking.
  • Text streams into view while you talk. Tap or click the waveform to hide it if it distracts you.
  • When you stop, review and edit the transcript, then send it as a normal text message.
  • OpenAI has not published a new maximum recording length.

Why it matters

For everyday users, dictation is one of the most used shortcuts in ChatGPT, especially on phones. Seeing text live makes long prompts much less of a gamble.

For professionals, removing the cap makes ChatGPT more usable for dictating drafts, notes and briefs, where recordings often run longer than ten minutes.

For the speech-to-text market, a free, real-time option built into ChatGPT raises the bar for standalone dictation apps. Open models are also improving fast, such as the open-weight Rumble-2-Flash speech recognition model released this week.

What’s still open

  • Whether there is any new practical limit on recording length.
  • How well live transcription handles languages other than English and noisy environments.
  • Whether the same streaming engine comes to the API for developers.

Sources

  • OpenAI, ChatGPT release notes, “Live Dictation in ChatGPT,” 18 November 2026
  • ChatGPT on X, 18 November 2026
  • Atty Eleti on X, 18 November 2026

i10x Perspective

Small interface changes often matter more than new models. Live text removes the anxious wait at the end of a long voice prompt, and dropping the time limit turns dictation into a real writing tool. More from ChatGPT and our Browsers & Assistants section.

Christopher Ort

Christopher Ort

Christopher Ort handles SEO, AI agent workflows and PR & partnerships at i10x.ai, the all-in-one AI workspace built by i10X Global Pte. Ltd. in Singapore. He is a graduate of Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU). He is the author of “The Style Penalty”, a study published on SSRN that analyses 1,576 evaluation decisions by four frontier AI models to show how AI-written CVs fare in automated hiring screens; the research has been covered by heise online and t3n.

Timeline

What happened before

  1. Release

    GPT-6.1 Terra

    1.05M context · $1 / $5 per 1M tokens

    openai.com (source, opens in a new tab)
  2. Release

    GPT-6.1 Sol

    1.05M context · $2 / $10 per 1M tokens

    developers.openai.com (source, opens in a new tab)
  3. Release

    GPT-6 Sol & Luna

    1.05M context · $2 / $10 per 1M tokens

    Two GPT-6 reasoning models. Price shown is GPT-6 Sol (≤272K prompt); GPT-6 Luna costs $0.10 / $0.50.

    developers.openai.com (source, opens in a new tab)
  4. Release

    GPT-6 Astra

    1.05M context · $10 / $50 per 1M tokens

    OpenAI's most capable model for reasoning, coding, computer use and research.

    openai.com (source, opens in a new tab)
  5. Release

    GPT-5.6

    1.05M context · $4 / $20 per 1M tokens

    GPT-5.6 family: Sol (frontier), Terra (balanced) and Luna (high-volume).

    developers.openai.com (source, opens in a new tab)
Full ChatGPT timeline ›

Editor's picks

  1. Circus Brings Its CA-1 Robot Kitchen Into a Hospital, Serving Fresh Meals Around the Clock at Schoen Clinic Munich
  2. Metano Emerges From Stealth With a $12 Million Seed to Secure AI Agents Across Endpoints, Cloud and SaaS
  3. IEA World Energy Outlook 2026: AI Boom Doubles Data Center Electricity Use to 1,000 TWh by 2030
  4. Know Your Agent: Singapore’s MAS Plans a Protocol for AI Agents in Banking and Warns Users Not to Share Their Logins
  5. AI coding agents targeted by self-spreading GitHub malware campaign