AI

ChatGPT Live Dictation Now Transcribes as You Speak

What Happened

OpenAI has added Live Dictation to ChatGPT, a change that lets spoken prompts appear as text while the user is still talking instead of only after the recording ends, according to The Verge. OpenAI is rolling the feature out globally across all ChatGPT plans, with availability covering web, desktop, iOS, and Android.

The core workflow is simple: while dictating, users can tap or click the audio waveform to show the live transcript, and once they finish, they can edit that transcript in the prompt box. OpenAI also said dictation is becoming faster and “more reliable”.

There is also a practical limit change here: OpenAI removed the earlier 10-minute recording cap on ChatGPT voice dictation. Release notes quoted by 9to5Mac similarly say that speech turns into text as the user talks and that the waveform control can be used to show or hide text as they go.

The Real UX Shift

This looks like a small interface tweak, but it changes how ChatGPT voice input feels in practice. Before, users had to wait for the full recording to finish before seeing the final text. Now the model surfaces the prompt as it forms, which makes dictation more transparent and easier to correct quickly, based on the behavior described by The Verge and the release notes cited by 9to5Mac.

Live Dictation Breakdown

Streaming Prompt Text

OpenAI’s update means ChatGPT now transcribes audio prompts as the user speaks, rather than waiting for recording to stop. OpenAI also says the experience is faster and more reliable.

Workflow:

Speak into ChatGPT → tap or click the waveform to reveal live text → finish dictation → edit the transcript in the prompt box before sending.

Global Rollout Across Plans

The feature is rolling out globally on all ChatGPT plans. Coverage listed in reporting includes web, desktop, iOS, and Android, which matches the release-note summary quoted by 9to5Mac.

Separate From Record Mode

The Verge notes that OpenAI’s separate record mode for transcribing voice notes and meetings is still in beta and remains limited to certain paid or workspace tiers. The same report also says this Live Dictation launch follows OpenAI’s earlier overhaul of ChatGPT’s two-way conversational voice mode.

Quick Feature Check

DailyOps Insight

Why it matters: This is less about a flashy new AI model and more about reducing friction in one of the most common AI workflows: getting a thought into the prompt box. By making dictation visible in real time and removing the old recording cap, OpenAI is smoothing out input latency and giving users more confidence in what ChatGPT actually heard, based on reporting from The Verge and 9to5Mac.

The takeaway: For users who rely on voice prompts, the meaningful upgrade is not just that ChatGPT listens. It now shows its work while listening.

Sources