Skip to content

Tools and accounts · Voice & dictation

Speak. Refine. Send.

Understand voice availability, permissions, and where your transcript goes.

Current availability

SwarmPane Voice comes with SwarmPane Pro and works in the Mac app (macOS 13 or later) and, from 1.3.16, the Windows app, where Ctrl+Shift+Space starts it while SwarmPane is in front. Starter has no voice: the voice controls and the shortcut show a short note with Upgrade instead of recording. The Linux app doesn’t include voice yet; it is next. Offline Whisper and Voice Anywhere are macOS only for now. The voice example on the website is simulated and does not use your microphone.

Dictate into the right place

  1. Click into the composer or terminal pane that should receive the words.
  2. Press ⌘⇧Space (Command + Shift + Space, the default shortcut) or use the voice control, and speak. A small pill shows that SwarmPane is listening, over full-screen apps too.
  3. Allow microphone access when macOS asks.
  4. Press ⌘⇧Space again to insert the words into the composer or terminal that’s active in SwarmPane, or Esc to cancel. Both work while you’re in another app.
  5. Review the draft and edit names, commands, and punctuation before sending. In a terminal the words are typed on the input line and never run.

Cloud voice and Whisper on your Mac

Cloud voice is the default: Deepgram Nova-3 transcribes each recording on Cloudflare Workers AI (Whisper large v3 turbo takes any clip Nova-3 is unsure of), and neither the audio nor the text is stored. It comes with SwarmPane Pro, with 1,500 minutes of cloud voice a month, about 25 hours (300 during the 5-day trial). It detects the language you speak.

Whisper on your Mac works offline and uses none of those minutes. Download Whisper Base (148 MB) or Whisper Turbo (574 MB) once in Settings → Voice & Dictation, and install whisper.cpp (brew install whisper.cpp). On your Mac, SwarmPane runs Whisper in English for now.

When cloud voice can’t be used, for example offline, signed out or out of minutes, dictation runs with Whisper on your Mac and the app says why before recording. If neither can run, dictation does not start: the app says what is missing and offers the fix.

Cloud voice takes up to 2 minutes at a time. With Whisper set up, a longer recording is transcribed on your Mac instead. Without it, the dictation stops at 2 minutes and what you said is transcribed; dictate again to add more.

Voice Anywhere (Labs)

Voice Anywhere types your words into the Mac app in front instead of SwarmPane. It comes with Pro, and it’s in Labs and off by default: turn it on in Settings → Labs, then turn on SwarmPane Voice in System Settings → Privacy & Security → Accessibility. It never types into password fields and never presses Return. ⌥ ⇧ ⌘ Space speaks commands to SwarmPane, and anything that spends needs a click.

Need a hand with your setup?TroubleshootingContact