Skip to content

README.md

Latest commit

 

History

History
34 lines (23 loc) · 1.23 KB

README.md

File metadata and controls

34 lines (23 loc) · 1.23 KB

Offline Voice Assistant for Linux

Private, on-device voice assistant built with local AI models. Runs entirely offline after initial setup—no cloud, no data sharing.

Features

  • Real-time speech-to-text: Fast dictation using distilled Whisper.
  • Local text generation: Contextual responses powered by Qwen2.5-0.5B-Instruct.
  • Voice-controlled automation: Launch apps and execute workflows via spoken commands.
  • Text summarization: Instantly summarize selected text with one hotkey.
  • Seamless desktop integration: Auto-types output directly into active windows using xdotool.

Built for speed and privacy on Linux desktops. Lightweight models run efficiently on consumer hardware (CPU/GPU).

Experience fluid, Mac-like voice interaction—completely local and under your control.

Installation

pip install torch transformers sounddevice scipy pynput
sudo apt install xdotool xclip

Usage

Run with python voice_assistant.py. Models auto-download on first launch (internet required once).

Hotkeys:

  • Alt+1: Dictation mode
  • Alt+2: Generate response from speech
  • Alt+4: Launch apps by voice
  • Alt+5: Summarize selected text

Designed and built by a GenAI engineer focused on practical, private Linux tools.