Dictation
Dictate instead of typing: the mic in the chat composer turns your speech into the prompt live. Where to find it, how it works and what to do if it fails.
Dictation writes your prompt from speech. Your words appear in the composer as you talk. Nothing is saved as a transcript or an asset, it costs no credits and works on every plan.
The mic ("Start dictation") replaces the send button when a chat composer is empty: on the workspace home, in any chat and on Social radar. The generate composers on Image, Video and Canvas don't have it.
Dictate a prompt
Select the mic
Allow microphone access the first time. A recording bar appears with ✕, a red dot and a timer.
Talk
Text appears as you speak and is finalized when you pause. It is added to anything already in the composer.
Stop or cancel
Select stop or press Enter to keep the text. Select ✕ or press Escape to discard everything from this recording; there's no undo.
Review and send
Dictation never sends. Edit the text if needed, then send.
Limits
- One dictation lasts at most five minutes, then stops and keeps the text. For longer audio, make a transcript.
- Each person can start six dictations a minute and sixty an hour.
- It needs a microphone and a live connection. If the connection drops, text already added stays.
If it doesn't start
| Message | What to do |
|---|---|
| "Microphone access is blocked. Allow it in your browser settings" | Allow the microphone for this site, then select the mic again |
| "No microphone found. Connect one and try again" | Connect an input device |
| "Dictation isn't supported in this browser. Type your message instead" | Use a current desktop or mobile browser |
| "An error occurred while starting dictation. Try again" | Try again; you may have hit the rate limit above |
| "An error occurred during dictation. Try again." | The session dropped. Select the mic again |