Chirkut: Voice-Enabled Chrome Extension

Overview

Chirkut is a voice-enabled Chrome extension that listens for the activation word "chirkut" and processes voice commands and performs various browser actions such as tab navigation, webpage interaction, content summarization, content creation and a lot more.

Demo

Chirkutv1.mp4

Features

• Voice Activation: Say "chirkut" to activate the assistant.
• Dynamic Command Execution:
  • Switching tabs and closing.
  • Scroll the webpage.
  • Interact with webpage elements.
  • Open specific websites.
  • Control media playback.
  • Perform Google searches.
  • Summarize or analyze webpage content.
  • Post generation or content creation(text).

Installation

• Clone the repository:

git clone https://github.com/KrishavRajSingh/chirkut.git

• Copy enviroments variable from .env.example and paste it in .env file. Add your Gemini Api Key in .env

• Open Chrome and navigate to chrome://extensions/.

• Enable Developer mode in the top-right corner.

• Click Load unpacked and select the project directory and then go to build/chrome-mv3-dev.

• This will open a new pinned tab which will start listening your voice.

Usage

Open any webpage and ensure your pinned tab is listening your voice. Say "chirkut" to activate the extension. Give voice commands such as:

• Switch to the next browser tab.
• Scroll down or go to bottom part of the page.
• Open YouTube in a new tab.
• Click balle-balle-te-shaba-shaba(youtube video element).
• Read the screen and tell me music videos present on the screen The assistant processes your commands and responds accordingly.