Dimmy — your voice, made powerful
Dimmy is a native, offline-first, privacy-first voice app for macOS and Windows (Linux from source). It turns your voice into a power tool with three capabilities: dictation, command mode, and meeting recaps. Open-core (AGPL-3.0), no Electron, tiny native binaries. Built by two developers in Milano.
Dictation — speak, and clean text lands where your cursor is
Press a hotkey, talk, release. Dimmy transcribes locally with Whisper and an optional AI enhance pass cleans up punctuation, grammar, and tone (13 styles). It pastes into any app: editors, browsers, chat, terminals. Up to 99 languages, fully offline by default, around 0.4s round-trip on an M1.
Command Mode — say what to do, and it happens in place
Flip the pill into command mode and your speech becomes an instruction. Select text and say "make it more formal", "translate to English", or "turn this into bullets", and Dimmy rewrites the selection right where it sits. With nothing selected, your words become a writing request at the cursor. Windows and macOS.
Meetings — record the call, keep the recap
A dedicated Dimmy Meeting window records both sides of a call (microphone + system audio), transcribes live, and on stop writes a structured recap: TL;DR, key decisions, action items, open questions, with timestamps. A meeting-type classifier tunes the sections, and a consent gate appears every session. The audio and transcript stay on your machine and are full-text searchable. Export a recap to Notion, a synced folder (Obsidian, Drive, Dropbox), or Claude Desktop, and the dimmy-mcp bridge lets Claude list, read, search, and save recaps.
Why Dimmy
- Offline & private: in local mode no audio leaves your device, and audio is never written to disk.
- Open source: full AGPL-3.0 source on GitHub; paid plans sell convenience, not features.
- Native, not Electron: WinUI 3 on Windows, SwiftUI on macOS, a shared Rust core.
- Three tools in one: dictation, voice commands, and meeting recaps in a single app.
- Works everywhere: 100+ apps, pasted via the clipboard, no per-app plugin.
Pricing
- Free — €0 forever, build from source (same code as Pro, no feature gates).
- Pro — €4.99/month or €39/year: signed prebuilt binary, in-app auto-update, beta access, human email support.
- Lifetime — €99 one-time. 14-day free trial, 30-day refund, students 40% off.
Frequently asked
Is Dimmy really free?
Yes, if you compile it yourself. The whole code is public on GitHub and builds on macOS, Windows and Linux. Pro and Lifetime sell the convenience: a signed prebuilt binary, one-click updates, early access, and a human on email.
What about my privacy?
Default mode is fully offline. Your audio never leaves your machine and is never written to disk. Cloud providers are opt-in per provider, and API keys are encrypted on device.
Can it handle meetings without a bot joining the call?
Yes. Dimmy records both sides locally (mic + system audio) with no bot and no third-party server, then writes a structured AI recap you keep and can search.
Help center · Source on GitHub · Privacy · llms.txt