This post is also available in: heעברית (Hebrew)

An innovative project involving advanced deep learning, voice forensic and big data technologies will enable early diagnosis of the Coronavirus. The platform finds specific patterns in voice, tone and other speaking sounds that relate to unique illnesses. 

The initiative, a collaboration between the Israeli company Voca.ai and Carnegie Mellon University, will be based on a vast database of recordings of human voices, including healthy and sick people. The participants will report details about their health condition, habits, genre, age, whether they are smokers, etc.

The development is based on the assumption that the coronavirus which affects the lungs and the breathing system has also influence on the vocal cords.

The data will be analyzed by employing deep learning technologies, that involve the training of the computer with a large number of samples so that eventually it would be able to recognize a sick person even before this person shows any symptoms.

The data will be open to anyone who can use it to identify virus patients, according to mako.co.il.