Dictating text

Speakly types where your cursor is. Hold the shortcut down and speak, or tap it, speak and tap again — either way the text lands in whatever app you were already using, whether that is Slack, Gmail or your editor.

How to dictate

1. Put your cursor where the text should go
2. Press the shortcut — the recording pill appears at the bottom of the screen
3. Speak normally
4. Press the shortcut again to finish — or let go, if you were holding it down
5. The transcribed text is pasted at your cursor

Cancelling a dictation

Changed your mind? Press esc while the recording is running and the dictation is cancelled: the audio is thrown away, nothing is transcribed or pasted, and Speakly is immediately ready for the next one. Esc is only intercepted while a dictation is actually recording — at any other time the key behaves normally in every app. An Agent Mode dictation is the exception: the agent bar has its own esc, which closes the bar instead.

What happens when you stop

• The audio goes to the transcription engine you picked (local or BYOK)
• If post-processing is on, the text passes through your AI model before being pasted
• The result is pasted into the window that had focus
• A copy is kept in your history, searchable later

Maximum length

A single recording runs up to 4 minutes 55 seconds. The timer on the pill shows how far along you are. For anything longer, dictate in chunks — each chunk is pasted as soon as it finishes.

If nothing happens when you press the shortcut

• You must be signed in — the shortcut is ignored without an active session
• Check microphone access under Settings → Permissions
• Pasting needs Accessibility permission on macOS
• Another program may have claimed the same combination — change it under Settings → General → Keyboard shortcut → View