Press Ctrl+Space, talk, press again. What you said is written out, cleaned
up and pasted into the window you were typing in. It runs on Linux, macOS and
Windows, on this machine by default.
![]() |
![]() |
| Nord (default) | Dracula | Classic light |
|---|---|---|
![]() |
![]() |
![]() |
-
Speech to text on your own computer by default, with the settings window downloading the model for you. Cleanup can run there too.
-
Or in the cloud, with OpenAI, OpenRouter or any OpenAI-compatible server through Custom server, with gateway presets and your own key.
-
A clean result. Filler words, restarts and missing punctuation are fixed, and so are names the model misheard:
raw ıı bugün şey kuber netis üzerinde çalışan servisleri güncelledim yani sonra grafanada bir panel açtım hani ve pay kut ile arayüzü şey bitirdim işte result Bugün Kubernetes üzerinde çalışan servisleri güncelledim. Sonra Grafana'da bir panel açtım ve PyQt ile arayüzü bitirdim. -
Writing profiles. A tone for each program, translation, and a dictionary of your names that can learn them as you correct.
-
Voice commands. Say a task to Claude Code, Codex or Antigravity and get the answer pasted back, or ask about the text you selected.
-
Meetings. Your microphone and the other side are recorded together, written out and turned into minutes. Not on Windows yet.
-
Audio and video files become text or
.srtsubtitles. -
History of every dictation, and a
diktecommand for everything the window does. -
Turkish and English interface.
Download it from the releases page: an AppImage for Linux, a
disk image for each Mac and a setup for Windows. The AppImage still needs your
distribution's packages for sound, the clipboard and pasting, listed in
docs/install.md. On a Mac the first launch is refused until
you press Open Anyway under System Settings → Privacy & Security. The Mac
shortcut is Ctrl+Option+Space.
To install from source on Arch:
sudo pacman -S --needed pipewire-audio wl-clipboard ydotool ffmpeg python-pyqt6
systemctl --user enable --now ydotool
./install.sh
dikteFedora, Ubuntu, macOS and Windows have their own steps in
docs/install.md. On KDE the shortcuts start working after
your next login; until then, turn on the built-in listener under Settings →
Shortcuts, which needs your user in the input group.
| What | How |
|---|---|
| Start / stop recording | Ctrl+Space, or click the tray icon |
| Pause / resume the recording | Tray menu, dikte pause, or a key you set |
| Discard the recording | Ctrl+Alt+Space, tray menu, or dikte cancel |
| Speak a command to an agent | Tray menu → Ask Claude, or dikte ask |
| Start / end a meeting | Tray menu → Record a meeting, or dikte meeting |
| Settings | Tray menu → Settings, or dikte settings |
| Look for a newer version | General tab → Check now, or dikte update |
| Reload after an update | Tray menu → Restart, or dikte restart |
| Quit | Tray menu → Quit, or dikte quit |
How each part works, and where the code for it lives, is in docs/how-it-works.md.
GPL-3.0, see LICENSE.





