Researchers in the Predi-Covid study led by Dr Guy Fagherazzi of the Luxembourg Institute of Health have shown that it is possible to monitor the progression of Covid-19 using voice evidence. (Photo: Shutterstock)

Researchers in the Predi-Covid study led by Dr Guy Fagherazzi of the Luxembourg Institute of Health have shown that it is possible to monitor the progression of Covid-19 using voice evidence. (Photo: Shutterstock)

The Luxembourg Institute of Health (LIH) has demonstrated, for the first time, that a voice recording of people with Covid-19 can be used to monitor the progression of the disease's symptoms.

Monitoring the progress of Covid-19 by voice is now possible according to researchers in the Predi-Covid study led by Dr Guy Fagherazzi of the Luxembourg Institute of Health (LIH).

His team is the first to use voice recordings of speech--not coughing or breathing--recorded by different devices in a natural environment to identify Covid-19-related symptoms.

Two years ago, the LIH called for volunteers to participate in a five-minute survey to collect voice and cough samples from a large panel of the population. To begin with, the researchers routinely analysed and collected voice data from the participants, 279 people divided into two groups: the symptomatic and the asymptomatic. The auditory characteristics of each group were then compared and processed by an artificial intelligence model to predict the patients' symptomatic status.

"In this way, the researchers were able to identify a voice biomarker that can be used to accurately monitor the progress of the disease in asymptomatic and symptomatic individuals," explained Fagherazzi.

Anticipating the evolution of long Covid

While the study may seem a little surprising, being able to monitor the progression of disease symptoms by voice is a significant advance in the age of teleconsultations. In the near future, health professionals will be able to use this new technology to screen their patients and monitor the progression of their symptoms via remote control, using inexpensive and non-intrusive tools such as a smartphone. It could also prove to be a practical solution to monitoring the symptoms of people with 'long' Covid to anticipate their progression.

"Such a voice biomarker could well be integrated into future remote monitoring solutions, our digital devices or clinical practice. This voice biomarker offers an easy and non-intrusive way of collecting data that can be done at home. This could profoundly change the way patients are monitored and treated, and provide a long-awaited answer to the problem of overloaded healthcare systems," said Fagherazzi.

This story was first published in French on . It has been translated and edited for Delano.