shushing face

speak naturally, get polished text.

⚠️ Early-stage project. Expect rough edges, breaking changes, and missing features. Not recommended for production use yet.

shushing face helps you work more efficiently by leveraging the fact that speaking is generally faster than typing. It operates in the background, allowing you to speak freely, and with a simple press or shortcut, it transcribes and cleans up your spoken words into clear, professional text, making it ideal for emails, reports, and other workplace communication.

Supported Platforms

OS	Version	Desktop	Status
Pop!_OS	24.04 LTS	COSMIC (Wayland)	Tested
Windows	10 / 11	—	Tested
Ubuntu	24.04 LTS	GNOME (Wayland/X11)	Expected to work
Ubuntu	22.04 LTS	GNOME (X11/Wayland)	Expected to work
Fedora	40+	GNOME (Wayland)	Expected to work

Install

From a release artifact

Download the latest build from releases:

• Pop!_OS / Ubuntu / Debian: sudo dpkg -i shushingface_*.deb
• Other Linux: extract shushingface-*.tar.gz and copy shushingface onto your PATH
• Windows: run the NSIS installer

From source

just doctor      # report missing dependencies
just bootstrap   # install missing dependencies (use --yes to skip prompts)
just dev         # run in dev mode
just install     # build + install for the current user

The same three commands work on Linux and Windows. Install locations follow each OS's convention:

• Linux: binary under $HOME/.local/bin/ (override with PREFIX=...), desktop entry + icon under $HOME/.local/share/
• Windows: binary under %LOCALAPPDATA%\Programs\shushingface\ plus a Start Menu shortcut

Runtime files also live in OS-standard locations:

	Linux	Windows
Config	$XDG_CONFIG_HOME/shushingface/config.json (default ~/.config)	%APPDATA%\shushingface\config.json
Logs + history DB	$XDG_STATE_HOME/shushingface/ (default ~/.local/state)	%LOCALAPPDATA%\shushingface\

Packaging (maintainers)

just package produces installable artifacts for the current OS:

• Linux: dist/shushingface-<version>-linux-amd64.tar.gz and dist/shushingface_<version>_amd64.deb
• Windows: NSIS installer in dist/

Usage

1. Launch shushing face
2. Set up an AI provider (Groq is free and fast)
3. Set your global shortcut in Settings → General → Recording (on Linux, just install also registers a system-wide Super+Ctrl+B binding)
4. Press the shortcut to start recording, press again to stop
5. Refined text is typed where your cursor is

Screenshots

Onboarding — connect a provider and pick a refinement style

Main views — record, browse history, test prompts in the playground

Settings — connections, models, rules, and general preferences