Face Recognition Authenticated Voice Assistant System for the disabled

Most of the visually challenged, handicapped, and elderly persons are in social isolation, voice assistant system could help them to overcome their difficulties and work independently instead of depending on others. In this work, a system in python is built to encourage blind, elder, and handicapped people to keep them engaged and work independently. Face Recognition system developed with OpenCV is used as the authentication mechanism. Authenticated users can use the virtual assistant system where instead of the conventional keyboard and mouse, voice commands are used to get their job done. This system takes voice command as input request through a microphone using speech recognition module and according to the request it does either web automation with selenium, get any information from a Wikipedia article through web scrapping, perform machine reading comprehension with the BERT model, request for the current time using time module, or control inbuilt LED of Arduino board through serial communication. The contents in the screen are converted into speech through the text to speech library. Response to the request by the user is given in the form of speech.