Researchers at the University of Cambridge have developed a new app that will collect voice, breathing and coughing sounds, which will enable machine learning algorithms to determine whether a person is likely to be infected with COVID-19.
The app has been launched as a web app for Chrome and Firefox browsers and is expected to be available on Android and iOS soon.
Since COVID-19 is a respiratory condition, affected patients may have specific voice, breathing and cough sounds. A large, crowdsourced data of such sounds could potentially facilitate automatic detection of the condition.
Apart from the basic demographic and clinical user data, the app collects samples of spoken voice, breathing and coughing using the phone’s microphone. The app will also inquire if the users have tested positive for COVID-19 and collect one coarse grain location sample. The data will be exclusively used for research purposes.
In addition to detection, the dataset is expected to provide insights on disease progression, and correlate respiratory complications with medical history.
Professor Cecilia Mascolo who is leading the app's development, said: "There are very few large datasets of respiratory sounds, so to make better algorithms that could be used for early detection, we need as many samples from as many participants as we can get."
