Jev in production › Model and agent routing

Light-Whisper

Windows dictation app where, in Auto mode, Jev decides whether AI polish, screen context or web search is needed for a transcript before the LLM runs; optional Jev checks also flag meaning changes after polishing.

Open on GitHub ↗

The source publishes no measured number.
Use
Model and agent routing
Industry
Productivity
Form
Open-source tool
Stage
In production
Plugs into
Windows
Listed
2026-09-20
Found via
github
Repository
sypsyp97/light-whisper
Stars
76
Forks
12
Last push
2026-11-17
Language
Rust
License
GPL-3.0

The README opens with

Local and cloud speech-to-text for Windows. Speak with a hotkey. Send the text to your active app.

Download for Windows   ·   Start   ·   简体中文

1. Install. Download the Windows 10/11 x64 x64-setup.exe from Releases. Start the installer. The installer includes the recognition runtime and its dependencies. 2. Select an engine. Open Settings. Select your microphone. Select a recognition engine. If the selected local model is missing, select Start Download on the main screen. For cloud recognition, enter your API key and region. 3. Speak. Put the text cursor in the target app. Hold F2 . Speak. Release F2 to send the text.

Badge

For the project's own README, linking back here:

Listed in Jev in production

Also used for model and agent routing