Key TakeawaysResearchers are evaluating a new COVID test that would detect the virus based on voice and sound.The test listens to how a person composes a sentence, in addition to their coughing and breathing sounds, to determine if they are COVID-19 negative or positive.The test is still under study and is not available to the public, but may represent a breakthrough in virus detection.

Key Takeaways

Researchers are evaluating a new COVID test that would detect the virus based on voice and sound.The test listens to how a person composes a sentence, in addition to their coughing and breathing sounds, to determine if they are COVID-19 negative or positive.The test is still under study and is not available to the public, but may represent a breakthrough in virus detection.

Researchers in the Netherlands are studying ifa mobile appcould detect COVID-19 through changes in a user’s voice.COVID-19 typically targets the upper respiratory system andthelarynx, commonly known as the voice box, and it can affect the way people sound.

The team built an artificial intelligence model based on data from University of Cambridge’sCOVID-19 Sounds App, which has over 800 audio samples from over 4,000 health participants and 308 people who tested positive for COVID-19. These samples included recordings of the participants’ coughing, breathing, and reading sentences out loud.

“Mild or severe changes in one’s voice can be a sign of a variety of diseases, making vocal biomarkers a noninvasive tool to monitor patients, grade the severity, the stages of a disease or for drug development,” the researchers wrote.

The analysis tool can distinguish loudness, power, and variation over time in a person’s voice. So far, the AI model is found to be 89% accurate in detecting the virus. This could be an effective, non-invasive COVID testing method if it becomes available in the future.

“These promising results suggest that simple voice recordings and fine-tuned AI algorithms can potentially achieve high precision in determining which patients have COVID-19 infection,”saidWafaa Aljbawi, a coauthor of the study and a researcher at the Institute of Data Science at Maastricht University in the Netherlands.

Vocal tests have been used to diagnose other conditions before. Some studies have shown that app-based audio tests can diagnose Parkinson’s disease with up to 99% accuracy.

When it comes to COVID-19, patients may have symptoms like a sore throat, a cough, or shortness of breath. Excessive coughing may lead to a hushed or raspy voice, and shortness of breath may make someone pause more often to finish saying a sentence.

What Is Dyspnea?

“Such tests can be provided at no cost and are simple to interpret. Moreover, they enable remote, virtual testing and have a turnaround time of less than a minute,” Aljbawi said. “They could be used, for example, at the entry points for large gatherings, enabling rapid screening of the population.”

It might take a while before a COVID voice test becomes available for public use. For now, the researchers said they need to test the tool with more samples to confirm its accuracy. More than 50,000 audio sampleshave been collectedin the database so far.

What This Means For You

The information in this article is current as of the date listed, which means newer information may be available when you read this. For the most recent updates on COVID-19, visit ourcoronavirus news page.

2 SourcesVerywell Health uses only high-quality sources, including peer-reviewed studies, to support the facts within our articles. Read oureditorial processto learn more about how we fact-check and keep our content accurate, reliable, and trustworthy.Aljbawi W, Simmons SO, Urovi V.Developing a multi-variate prediction model for the detection of COVID-19 from crowd-sourced respiratory voice data.arXiv. Preprint posted online September 8, 2022. doi:10.48550/arXiv.2209.03727Singh S, Xu W.Robust detection of Parkinson’s disease using harvested smartphone voice data: a telemedicine approach.Telemed J E Health. 2020;26(3):327-334. doi:10.1089/tmj.2018.0271

2 Sources

Verywell Health uses only high-quality sources, including peer-reviewed studies, to support the facts within our articles. Read oureditorial processto learn more about how we fact-check and keep our content accurate, reliable, and trustworthy.Aljbawi W, Simmons SO, Urovi V.Developing a multi-variate prediction model for the detection of COVID-19 from crowd-sourced respiratory voice data.arXiv. Preprint posted online September 8, 2022. doi:10.48550/arXiv.2209.03727Singh S, Xu W.Robust detection of Parkinson’s disease using harvested smartphone voice data: a telemedicine approach.Telemed J E Health. 2020;26(3):327-334. doi:10.1089/tmj.2018.0271

Verywell Health uses only high-quality sources, including peer-reviewed studies, to support the facts within our articles. Read oureditorial processto learn more about how we fact-check and keep our content accurate, reliable, and trustworthy.

Aljbawi W, Simmons SO, Urovi V.Developing a multi-variate prediction model for the detection of COVID-19 from crowd-sourced respiratory voice data.arXiv. Preprint posted online September 8, 2022. doi:10.48550/arXiv.2209.03727Singh S, Xu W.Robust detection of Parkinson’s disease using harvested smartphone voice data: a telemedicine approach.Telemed J E Health. 2020;26(3):327-334. doi:10.1089/tmj.2018.0271

Aljbawi W, Simmons SO, Urovi V.Developing a multi-variate prediction model for the detection of COVID-19 from crowd-sourced respiratory voice data.arXiv. Preprint posted online September 8, 2022. doi:10.48550/arXiv.2209.03727

Singh S, Xu W.Robust detection of Parkinson’s disease using harvested smartphone voice data: a telemedicine approach.Telemed J E Health. 2020;26(3):327-334. doi:10.1089/tmj.2018.0271

Meet Our Medical Expert Board

Share Feedback

Was this page helpful?Thanks for your feedback!What is your feedback?OtherHelpfulReport an ErrorSubmit

Was this page helpful?

Thanks for your feedback!

What is your feedback?OtherHelpfulReport an ErrorSubmit

What is your feedback?