Skip to Content

Installation

Download a build, or run Glint from source.

You need a webcam and a free API key from physiology.presagetech.com. To look around first, you need neither: choose Try a recorded session on the first screen.

Download a build

Builds are on the Releases page.

chmod +x Glint-*.AppImage
./Glint-*.AppImage

If it says AppImages require FUSE, which is normal on Arch and Ubuntu 22.04 or newer, run it like this or install libfuse2:

APPIMAGE_EXTRACT_AND_RUN=1 ./Glint-*.AppImage

The SDK needs glibc 2.35 or newer.

Give an installed app your key

Set PRESAGE_API_KEY in your environment, or create a file named .env in the app's data folder containing PRESAGE_API_KEY=your-key:

SystemFolder
Linux~/.config/Glint/
Windows%APPDATA%\Glint\

The key is never bundled into a build.

1. EnvironmentPRESAGE_API_KEYset in your shellnotset2. .env by the sourcesUsed when you runnpm startnotset3. .env in the data folderUsed by aninstalled app
Glint stops at the first place that has the key. If none does, the camera will not start and you see “No Presage API key”.

Run from source

Needs Node 20 or newer. npm install downloads the SDK runtime for every platform, which is a few hundred MB.

git clone https://github.com/yashksaini-coder/glint
cd glint
npm install
cp .env.example .env   # then put your key in PRESAGE_API_KEY
npm start

Linux speech

Most Linux systems give the app no built-in voice, so Glint falls back to espeak-ng. Without it, your text is still copied, but nothing is said aloud.

sudo pacman -S espeak-ng     # Arch
sudo apt install espeak-ng   # Debian, Ubuntu

Next: type your first word.

Scope

What has been tested, what has not, and what Glint does not claim.

Basic usage

Type and speak your first word, first with the keyboard and then with the camera.

On this page

Download a buildGive an installed app your keyRun from sourceLinux speech