Could your voice tell you have COVID-19? MIT scientists think so

Researchers found evidence of vocal biomarkers, or measurable indicators, of disease

covid-19 rep Representational image | Shutterstock

As the COVID-19 pandemic continues its ruthless march across the world, doctors and researchers are working on multiple measures to counter it. From fast-tracking vaccine trials to trying out new medicines to combat COVID-19, scientists are doing it all. However, the tasks of these COVID-19 warriors have been made more complicated by the fact that asymptomatic cases appear to be growing.

In June, researchers estimated that 40 to 45 per cent of all COVID-19 cases are asymptomatic, with carriers able to transmit the virus longer than 14 days. Most testing for COVID-19 is still predominantly based on the use of swabs to extract throat samples. With testing usually prioritised to examine people with symptoms, the risk of asymptomatic COVID-19 patients being excluded is ever present.

Earlier this week, a research team at the Massachusetts Institute of Technology published a study in IEEE Open Journal of Engineering in Medicine and Biology. The study by researchers at MIT's Lincoln Laboratory processed speech recordings of people infected with COVID-19, who had not begun showing symptoms. MIT News reported "researchers found evidence of vocal biomarkers, or measurable indicators, of the disease. These biomarkers stem from disruptions the infection causes in the movement of muscles across the respiratory, laryngeal, and articulatory systems".

Thomas Quatieri, a senior staff member in the Lincoln Laboratory's Human Health and Performance Systems Group, had been involved in research on finding vocal biomarkers of neurological disorders such as amyotrophic lateral sclerosis (ALS) and Parkinson's disease. "These diseases, and many others, change the brain's ability to turn thoughts into words, and those changes can be detected by processing speech signals," MIT News reported. Quatieri and his team wondered whether such biomarkers existed for COVID-19.

"When symptoms manifest, a person typically has difficulty breathing. Inflammation in the respiratory system affects the intensity with which air is exhaled when a person talks. This air interacts with hundreds of other potentially inflamed muscles on its journey to speech production. These interactions impact the loudness, pitch, steadiness and resonance of the voice—measurable qualities that form the basis of their biomarkers," MIT News explained.

The MIT researchers combed YouTube for celebrities who had given interviews when they were asymptomatic and compared their voice samples to videos taken before they tested positive for COVID-19.

The MIT team used algorithms to analyse vocal signals and picked up disruptions in a subject’s voice associated with changes in movements of larynx and muscles.

"They hypothesised that COVID-19 inflammation causes muscles... to become overly coupled, resulting in a less complex movement," MIT News reported. The researchers found a "decreased complexity of movement in the COVID-19 interviews as compared to the pre-COVID-19 interviews".

"These preliminary results hint that biomarkers derived from vocal system coordination can indicate the presence of COVID-19," MIT News reported. However, the researchers conceded more data is needed before coming to conclusions. The research team is now working on data from Carnegie Mellon University of voice samples from people who were COVID-19 positive.

The real significance of the MIT team's research is in the fact that a voice-based detection system for COVID-19 could be implemented on a much larger scale than current methods that rely on physical samples. The predominant approach would be integration with mobile apps.

"A partnership is under way with Satra Ghosh at the MIT McGovern Institute for Brain Research to integrate vocal screening for COVID-19 into its VoiceUp app, which was initially developed to study the link between voice and depression. A follow-on effort could add this vocal screening into the How We Feel app. This app asks users questions about their daily health status and demographics, with the aim to use these data to pinpoint hotspots and predict the percentage of people who have the disease in different regions of the country. Asking users to also submit a daily voice memo to screen for biomarkers of COVID-19 could potentially help scientists catch on to an outbreak," MIT News reported.

The team is also planning to broaden its research parameters, including understanding different recording environments, emotional state of subjects and any other illnesses that could cause vocal changes.