AI project aims to diagnose COVID-19 using voice analysis

covid 19 diagnosis diagnose ai artificial intelligence voice speech recognition
AI project aims to diagnose COVID-19 using voice analysis
Ryan is an editor at TechForge Media with over a decade of experience covering the latest technology and interviewing leading industry figures. He can often be sighted at tech conferences with a strong coffee in one hand and a laptop in the other. If it's geeky, he’s probably into it. Find him on Twitter: @Gadget_Ry

Researchers from Carnegie Mellon University are developing an AI-powered voice analysis system for diagnosing COVID-19.

Governments around the world are racing to obtain sufficient and effective testing kits to diagnose COVID-19. The current widely-used test requires a thin cotton swab to be put up the nasal cavity and reach to the back of the throat. It’s not a painful procedure, but it’s invasive and uncomfortable.

COVID-19 tests which only require a finger prick are starting to be rolled out but obtaining sufficient numbers of any test is proving difficult. On Tuesday, British cabinet minister Michael Gove said the UK was being hindered by the global shortage of chemical reagents needed for testing.

If the researchers from Carnegie Mellon are successful, a test that could be taken at home instantly could be rolled out. While it’s unlikely to ever be as accurate as a full test, it could help to prioritise where limited resources should be allocated and determine which households are more likely to be suffering from seasonal flu.

Speaking to Futurism, Benjamin Striner, a graduate working on the project, said: “I’ve seen a lot of competition for the cheapest, fastest diagnosis you can have.”

“And there are some pretty good ones that are actually really cheap and pretty accurate, but nothing’s ever going to be as cheap and as easy as speaking into a phone.”

Coronavirus is a respiratory illness and therefore affects breathing patterns and other vital parameters. The AI system analyses a person’s voice and provides a score on the likelihood that the individual has coronavirus based on markers observed from known sufferers.

The researchers are currently asking both healthy and infected people to share a recording of their voice to help improve the algorithm.

Yours truly has already submitted his voice. The process takes around five minutes and requires the following seven steps:

  • Submit basic demographic information.
  • Cough three times.
  • Say ‘a’ for as long as you can.
  • Say ‘o’ for as long as you can.
  • Say ‘e’ for as long as you can.
  • Count to 20.
  • Say the alphabet.

Be aware the app is still in its early stages and is not yet approved by agencies like the FDA or CDC. The app should also not be used as a substitute for a proper medical test or examination if you’re concerned you may have COVID-19.

If you’re interested, you can access the COVID Voice Detector here.

Interested in hearing industry leaders discuss subjects like this? Attend the co-located 5G Expo, IoT Tech Expo, Blockchain Expo, AI & Big Data Expo, and Cyber Security & Cloud Expo World Series with upcoming events in Silicon Valley, London, and Amsterdam.

Tags: , , , , , , , ,

View Comments
Leave a comment

One comment on “AI project aims to diagnose COVID-19 using voice analysis

Leave a Reply