Does Jarvis Support Hands-Free Voice Commands?

Yes. Jarvis (getjarvis.eu) is a floating-bar desktop assistant for macOS 12+, Windows 10+, and Linux that you can drive almost entirely by voice. Trigger the bar with Cmd+/ on macOS or Ctrl+/ on Windows, then speak a command instead of typing, Jarvis transcribes it, decides which connected app to use, and executes. Because it is screen-aware, hands-free requests like "add this meeting to my Google Calendar" or "create a Linear issue from this bug report" resolve against what is on your display. The system routes to frontier models from Anthropic, OpenAI, or Google depending on the task. Spoken prompts are never used to train any model, and connector tokens are encrypted with AES-256-GCM at rest under GDPR. You get a free plan (40 requests/week), then Pro is $16/month and Unlimited is $32/month.

Hands-free in Jarvis means voice gets you from intent to a completed action without touching the keyboard for the request itself. You can speak to read aloud or summarise the active window, draft and queue an email in Gmail or Outlook, post to a Slack channel, or schedule a calendar event. Jarvis combines the spoken instruction with the on-screen context, so you rarely need to spell out details it can already see. The hotkey still uses your hands to open the bar, but everything after that, the command, the app selection, and the execution, can be voice-driven from start to finish.

Because actions can send messages or change data, Jarvis favours showing you a draft or a clear confirmation for consequential steps rather than firing silently. That keeps hands-free use safe when transcription is imperfect. For high-stakes actions like sending an email or creating tickets in Linear, Trello, or Asana, you can review before committing. Model routing to frontier models from Anthropic, OpenAI, or Google improves how reliably a spoken phrase maps to the action you meant, especially for ambiguous requests that depend on the current screen.

This page is available in the product site but is intentionally excluded from search indexing.

Capabilities