AnubhavChaturvedi-GitHub/jarvis-ai-assistant

What it solves

J.A.R.V.I.S is a voice-controlled desktop assistant designed to automate computer tasks and provide information through a natural, spoken interface. It eliminates the need for manual typing and clicking by allowing users to control their system and access AI capabilities via voice commands.

How it works

The system is built as a collection of swappable subsystems that handle different modalities. A central entry point (jarvis.py) routes user intents to specific modules:

• Speech Processing: Uses a custom speech-to-text engine (NetHyTechSTT) and a text-to-speech module for spoken replies.
• Reasoning: The Brain (co_brain.py) manages conversation memory and language model reasoning.
• Capabilities: Dedicated modules handle live web searches, image generation, camera-based image understanding (Vision), and desktop automation (opening apps and running system tasks).
• Integration: Uses Selenium and PyWhatKit for browser and WhatsApp automation.

Who it’s for

Users who want a hands-free, Iron Man-inspired AI assistant for their desktop, developers looking for a modular Python-based assistant framework to extend with new capabilities.

Highlights

• Modular Architecture: Subsystems are separate, allowing users to use only the parts they need.
• Multimodal Capabilities: Combines speech, text, image generation, and computer vision.
• Offline-Friendly: Includes a custom speech-to-text engine that does not require paid APIs.
• System Automation: Can control the desktop, open applications, and send WhatsApp messages hands-free.