Skip to content

Lakshay-Sood/Jexi---Desktop-Assistant

Repository files navigation

Jexi - Desktop Voice Assistant

A desktop automation project with voice assistant support.

For a complete report on Jexi, follow this document: https://docs.google.com/document/d/16mDcuC6_itjg1_lE3AypzI8F12A_Ymi717PFZwu2nKU/edit?usp=sharing

Brief Steps involved in the working of Jexi are:

  1. Set energy threshold values to determine when to start/stop recording
  2. Record audio from an audio source (usually system's microphone)
  3. Send this audio clip to the Google Speech Recognition API which return a text
  4. Verify if this text corresponds to any pre-defined executable command
  5. If yes, execute the corresponding command. Else, do nothing.
  6. Repeat steps 2-6.

Requirements & Setup

  1. You need to have python 3 installed on your system.
  2. Clone/Download this repository to your local machine.
  3. Download the required packages by running the " pip install -r requirements.txt " command in the terminal at the location where you clone/download this repository.
  4. Voila! You are done.

Bringing Jexi to life

  1. Simply run the python script "main.py" in the terminal.
  2. You are ready to use it now!
  3. Optional: Check the dictionary (hashmap) at the beginning of the file to get to know the commands that Jexi currently supports.
  4. Optional: For Video demo, check out this drive folder: https://drive.google.com/drive/folders/1ZT2Sg4BZ8A212ORlAgedqnDRmzWH8npd?usp=sharing