Connect with us


Smartphone app accurately detects Covid-19 infection in people’s voices 2022



Smartphone app accurately detects Covid-19 infection in people's voices 2022

There is still COVID-19 around. People won’t have to wait in line for the COVID testing anymore, though. A new smartphone app was developed by academics from Maastricht University’s Institute of Data Science in the Netherlands. According to the researchers, using artificial intelligence, the app can precisely identify COVID-19 infection in people’s voices (AI).

The app, according to the development team, is less expensive, quicker, and simpler to use than a number of antigen testing methods. It implies all you need is a smartphone to determine whether you have the virus or not. This will be a ground-breaking app, especially for developing nations with little financial resources.

Researcher Wafaa Aljbawi from Maastricht University in the Netherlands’ Institute of Data Science said

The encouraging findings imply that straightforward voice recordings and well-tuned AI systems may be able to identify whether people are infected with COVID-19 with high precision. They also allow for remote and virtual testing and have a quick response time. For instance, they could be used, for instance, at the entrances to major gatherings to enable quick population screening.

Mel-spectrogram analysis is a voice analysis method that the researchers use.

The upper respiratory tract and vocal cords are frequently impacted by the COVID-19 infection, altering a person’s voice.

Aljbawi and her superiors made the decision to look into the viability of using AI to analyse speech in order to recognise COVID-19.

They used information from the crowdsourced COVID-19 Sounds App from the University of Cambridge, which includes 893 audio samples from 4,352 healthy and unhealthy subjects, of whom 308 had tested positive for Covid-19.

See also  Are the primitive building videos real?

Researchers use Mel Spectrogram Analysis to tell the difference between different aspects of the voice, such as loudness, power, and changes over time.

This AI model is correct 89 percent of the time.

On Monday, the researchers demonstrated the app at the International Congress of the European Respiratory Society in Barcelona, Spain. They claim that the AI model is correct 89 percent of the time.Contrarily, depending on the manufacturer, the accuracy of lateral flow testing varies greatly.

Researchers say that the new test will be useful at the entrances to big meetings because it will let people be quickly screened.

How does the new app function?

The app must first be installed on the users’ devices. The participants provide some basic demographic, medical, and smoking status information. Users will be prompted to record their breathing noises.

Coughing three times and taking three to five full breaths through their mouth are a couple of these. Additionally, it entails reading a brief sentence three times on the screen.

Thanks for reading!

Stay connected with postvines for more information!!

Copyright © 2022 - 2023 PostVines,All rights reserved.

You cannot copy content of this page