Jarvis – Linux Voice Assistant (Offline, Intent-Based AI)

Jarvis is a Linux-only AI desktop assistant that you can open instantly using keyboard shortcuts.
It supports voice + text commands, uses a local NLP model to understand user intent,
and responds using a local Edge-TTS server (open-source).

Built for speed, and full on-device control.

Full Demo (with voice)

Click here to watch the full demo

---

Features

Global Popup Launcher

• Double-press Ctrl → Jarvis appears anywhere on your screen.
• Type or speak commands instantly.

Voice Control

• Hold Ctrl + S → speak
• Release → Jarvis processes your speech
• Converts speech → text → NLP → action

Local NLP Intent Model

Understands commands such as:

• open_app
• search_web
• play_media
• scan_network
• get_info
• system_control
• ...and more.

Local TTS (Edge-TTS Server)

• Jarvis speaks responses using a local Edge-TTS server.
• Fully offline and open-source.
• Press Q anytime to stop the speech output.

Linux Only

Supports Linux desktops.

---

How to Use

Action	Shortcut
Open Jarvis popup	Ctrl (twice)
Voice command	Hold Ctrl + S, speak, release
Stop voice speaking	Press Q
Close popup	click anyware or Ctrl (twise)

Example Commands

• “Open VS Code”
• “Scan my open port”
• “Who is Elon Musk?”
• “What is my system condition”
• “Open my last workspace”
• “Play song”

---

Current Capabilities

• Weather
• WhatsApp message
• Open Instagram
• Open Facebook
• Open camera
• Wikipedia (Elon musk)
• Open Google
• Open YouTube
• Play song
• Volume (up/down)
• Today’s Schedule
• Purchase Options (Flipkart , Amazon)
• Suggestions
• Open my yesterday night workspace
• Ip address
• System conditions
• Internet speed
• Scan vulnerable ports
• Scan open ports

Architecture

Double Ctrl → Popup UI → (Optional: STT) → NLP Intent Model → Skill Router → Linux Action Module → Result (Text + Edge-TTS Voice)

Requirements

• Linux (Ubuntu / Arch / Fedora / etc.)
• Python 3.x
• Installed local Edge-TTS server

setup process will be updated soon