Several teams of artificial-intelligence researchers are working toward systems that could help spot the disease through people’s breathing, coughing and speech. They are collecting thousands of sound samples from people who have contracted Covid-19 on the hope that the signature respiratory problems that make the infections so dangerous could also help unlock the way to a faster diagnosis.

Tomas Teijeiro, a researcher in Switzerland, was inspired when doctors told his team that coronavirus patients had a distinctive dry cough. “They’d see a patient come in, they’d hear them coughing, and they’d say ‘OK, it will be Covid,’ ” he said…

“You’ve got these quick-off-the-mark computer science projects that maybe hog the attention and resources away from something more deserving,” said Emily Bender, a computational linguist at the University of Washington who studies the social impact of speech and language technologies. The voice projects may be well-intentioned, she said, but there is a risk that they do more harm than good in the real world.