OpenAI has introduced a Live Dictation feature for ChatGPT that converts speech to text while the user is still talking, instead of waiting for a recording to finish. The company said the capability is rolling out globally across all ChatGPT plans. It is available on the web, the desktop app, and the iOS and Android apps.

While dictation is active, users can tap or click the audio waveform to view the transcript ChatGPT is producing in real time. When they finish speaking, the text can be edited directly in the prompt box. OpenAI said the transcription is faster than before.

OpenAI also announced that it is lifting the 10-minute recording limit that previously applied to voice dictation in ChatGPT. The company said the feature is becoming more reliable. The changes apply to all ChatGPT plans.

The dictation update follows a reworking of ChatGPT's two-way conversational voice mode earlier in 2026, according to OpenAI. The company has been expanding the audio capabilities of ChatGPT over the past two years.

In 2025, OpenAI announced a record mode for transcribing voice notes and meetings. That feature remains in beta and is restricted to Plus, Enterprise, Edu, Business, and Pro workspaces. Users on other plans can still employ the prompt box dictation feature to have ChatGPT transcribe voice messages or convert them into bullet points.

The global rollout means US users on free and paid ChatGPT tiers have access to the live transcription tool on the same schedule as other markets. OpenAI has not said whether the feature will be extended to the record mode beta.

The announcement covers only the dictation changes. OpenAI did not provide a specific release date beyond saying the feature is rolling out now.

More software news from TechManNews.