Skip to main content
Point at things while you talk. During a hands-free dictation, Cmd + click (Mac) or Ctrl + click (Win) anything on screen. Vowen takes a screenshot of the area around your pointer. When the dictation ends, your text is pasted first and the screenshots follow it into the same field. This is useful when you dictate a prompt for ChatGPT, Claude or a coding agent and want to show what you are talking about. Screenshot capture is a Pro feature on macOS and Windows.

​
Turn it on

1

Open Experimental settings

Go to Settings › Experimental and turn on Capture screenshots while dictating.Settings › Experimental with Enable Connectors, Capture screenshots while dictating and Enhanced silence detection
2

Grant Screen Recording (macOS)

On macOS the row stays blurred until Vowen has Screen Recording permission, with the message Turn on Screen Recording for Vowen in System Settings. See Permissions.

​
Capture screenshots

1

Start a hands-free dictation

Press the hands-free shortcut, or a dictation shortcut set to Tap mode. Capture doesn’t work while you hold a push-to-talk shortcut, because the shortcut’s keys are already held down.
2

Cmd+click or Ctrl+click what you mean

Each click captures the area around your pointer. The indicator shows Screenshot added, with a count once you have more than one.The recording indicator with a Screenshot added (2) chip under it
3

Stop the dictation

Vowen pastes your text, then each screenshot in the order you took them.

​
How it behaves

  • Area. Each screenshot is a fixed box centred on your pointer, kept inside the display. The click itself still reaches the app, but the screenshot shows the screen as it was before the click.
  • Limit. Up to 6 screenshots per dictation. Further clicks are ignored.
  • No Auto Enter. A dictation that carries screenshots is never sent with Auto Enter, so you can check it before sending.
  • Clipboard. Images go in through the clipboard. With Restore clipboard after paste on, your clipboard is put back afterwards.
  • Terminal coding agents. On macOS, when you dictate into Claude Code or Codex in a terminal, the images are pasted with Ctrl + V, which is the key those tools use for images.
  • Privacy. Screenshots are kept in a temporary folder for the length of one dictation and deleted after they are pasted, or when you cancel. Vowen doesn’t upload them anywhere. They only go into the app you dictated into.
  • Command Mode uses its own region capture instead.

​
Direct insertion

Coming in the next update.
With Text Insertion Method set to Direct insertion, the text is typed and the screenshots are then pasted through the clipboard. In the current release, screenshots are dropped when direct insertion is on.