Speak naturally. Get perfect text.

Hold Fn, say anything, release — and polished text appears exactly where you need it. Powered by on-device Whisper and cloud AI.

  • Inbox on the move
  • Meetings, eyes up
  • Chat without friction
  • Ideas into drafts

One voice workflow, four surfaces

Desktop hotkey everywhere you type. Mobile keyboard wherever you chat.

macOS

System-wide Fn hotkey, local Whisper, auto-paste.

Available

Windows

Same hold-to-speak flow for Outlook, Slack, browsers, and more.

Available

iOS

Custom keyboard with tone modes, mic handoff, and undo.

Available

Android

Voice keyboard overlay with the same rephrase mode set as desktop.

Available

Four beats. Zero friction.

Private transcription on-device. Tone-aware refinement in the cloud. Paste where you already are.

  1. 01

    Hold or tap

    Desktop hotkey or mobile mic — no app switching to start speaking.

  2. 02

    Transcribe locally

    Whisper runs on your device. Audio stays with you.

  3. 03

    Match the mode

    Rephrase, Professional, Casual, Formal, GenZ, or Research/Task.

  4. 04

    Auto-insert

    Refined text lands in the active field — with undo on mobile.

Paste into the tools you already live in

Terminal, Slack, Outlook, Messages, Gmail — and every other text field on your system.

Six ways to shape what you said

Pick a mode once. Every dictation follows that voice until you change it.

You said

“um so basically i was thinking we should maybe like push the deadline back a week because the team is kinda swamped right now”

Rephrase

“I’d like to extend the project deadline by one week. The team is at capacity, and the extra time will help us deliver quality work.”

Rephrase cleans up natural speech into clearer, well-structured text without forcing a specific personality.

Get Voice Input

Desktop builds ship below. Mobile builds are available — reach out if you need TestFlight or Play access.

Desktop releases

Latest pinned. Older builds stay compact.

Loading latest release…

Voice audio is processed on-device. Only the text transcript is sent for AI refinement.