Offline, simple, system-wide speech-to-text for Linux using #openai 's Whisper.
Press a shortcut, speak, and the text is typed directly into the currently focused text field. Runs fully locally. No cloud, no API keys, no telemetry. Works on every Linux distro.
Git Repo: https://github.com/Kr-Adarsh/whisper-...
Like & subscribe if this helped. Star the repo if you found it useful. If you've any doubts or need clarifications, drop a comment — I’ll help.. (^^)
---
Quick install
1) Clone:
git clone https://github.com/Kr-Adarsh/whisper-...
cd whisper-dictate
chmod +x install.sh
2) Install (CPU default):
./install.sh
3) Optional: CUDA (only if you want GPU acceleration & have drivers):
./install.sh --cuda
4) Run test:
source .venv/bin/activate
./whisper_dictate.sh
press ESC to stop
5) Bind shortcut (GNOME):
Settings → Keyboard → Shortcuts → Add custom
Command: /path/to/whisper-dictate/whisper_dictate.sh
Shortcut: choose e.g. Ctrl+Alt+D
---
#linux #opensource #openai #whisper #productivity #offlineai #voicetyping #llm #voice #dev