Detecting COVID in a Human Voice With A.I.

Google+ Pinterest LinkedIn Tumblr +

News Medical & Life Sciences says researchers in Netherlands have discovered a way to identify COVID infections from people’s voices. They presented their finding at European Respiratory Society International Congress in Barcelona, Spain on Monday September 5, 2022. Now we think of it, detecting COVID in a human voice seems logical, given it is a respiratory disease.

Detecting COVID in a Human Voice Is Faster and Cheaper

Moreover, the Netherlands researchers claim their new method is more accurate than lateral flow / rapid antigen tests. And as an added bonus it is also cheaper, quicker and easier to use. This should therefore make it especially attractive in low-income countries, where brick and mortar tests can be a challenge to administer.

Wafaa Aljbawi is a researcher at the Institute of Data Science, Maastricht University in Netherlands. She assured the European Respiratory Society International Congress the AI version was accurate 89% of the time.

Whereas lateral flow testing results vary widely depending on brand, she explained. Moreover AI is more reliable in the case of asymptomatic cases too. The rationale behind detecting COVID in a human voice is also straightforward.

That’s because a COVID-19 infection usually affects the upper respiratory track and vocal cords. And this causes changes in an affected person’s voice.

The Maastricht team used information from University of Cambridge’s crowd-sourcing COVID-19 Sounds App for their research. This provided them with a database of 4,352 healthy and non-healthy participants, 308 of whom tested positive for COVID-19.

How the COVID-Testing AI App Works

The app first detects the user’s voice characteristics via their smartphone microphone. The AI then asks for basic information concerning demographics, medical history and smoking status. After that, it then asks them to cough three times, breathe deeply through their mouth three to five times, and read a short sentence on the screen three times.

Provisional results suggest simple voice recordings and fine-tuned AI algorithms should be able to accurately detect COVID-19 infected persons. Thus, this appears to be an amazing step forward using two of the commonest devices in the world, the smartphone and the battery.

Breaking News

Chengdu Where People Still Go Shopping Daily

Is Vaccine Fatigue Cooling Booster Interest?

Preview Image: Mobile Subscriptions per 100 People

Share.

About Author

Electricity has always intrigued me. I have been writing about batteries and energy storage for more than ten years, and have published over 4,000 articles on this website. During that time, I have researched developments across lead-acid, lithium-ion, sodium-ion, flow batteries, and emerging energy-storage technologies. My goal is to explain complex battery concepts in clear, practical language that anyone can understand. My writing career began unexpectedly after leaving the corporate world. What started as a search for a new direction gradually became a fascination with batteries, renewable energy, and the science that powers modern life. Writing may not have made me wealthy, but it has given me the opportunity to explore an industry that continues to evolve in remarkable ways.

Leave A Reply