Researchers struggle with COVID-19 voice detection apps - report

Efforts to create apps that listen to users’ coughs and voices to predict whether they have the COVID-19 coronavirus have hit speed bumps, according to a press report.

The BBC reported that two universities are trying to develop software that uses machine learning to aim to identify those who are likely to have the disease, and distinguish from those who have colds or the flu.

In the UK, the BBC reported that a project by Cambridge University tries to keep volunteers anonymous, which is limiting its work.

An independent effort by the US-based Carnegie Mellon University asks users to register themselves but has had to go offline.

Both teams say the software does not replace the need for other medical tests, which would be used to produce a more accurate diagnosis.

Cambridge University launched its project on Tuesday where members of the public are invited to breathe and cough into a computer’s microphone.

They are asked details about their age, gender, approximate location, and whether they have tested positive for the coronavirus.

People using the app are asked to read the following phrase three times: “I hope my data can help to manage the virus pandemic.”

The team behind the app say that it is not yet available on smartphones because Google and Apple have yet to accept it because of tougher rules on COVID-19 related apps.

Getting the app validated for a smartphone is key because it would allow for follow-up tests, something that is difficult on a web app without compromising confidentiality, the BBC reported.

Meanwhile the Carnegie Mellon’s team Covid Voice Detector is built on other voice-profiling work done at the university in Pittsburgh.

The app briefly went live on 30th March and asked users to cough, record vowel sounds and recite the alphabet, as well as provide details about themselves.

At the end of the assessment the app gives an indication of how likely they were to have COVID-19.

But researchers realised a rethink was required because of the ethical concerns about a tentative diagnosis of the disease.

Researcher Dr Rita Singh told the BBC: “It doesn't matter how many disclaimers you put up there - how clearly you tell people that this has not been medically validated - some people will take the machine as the word of God.”

False positives could waste precious resources and missing a person who has developed the disease could have fatal consequences, according to Singh.

“So, we have very little room for error either way, and are deliberating on how to present the results so that these risks vanish.”

The Carnegie Mellon tool may be relaunched within a few days using a system where people register with email addresses instead of their names.

It will also include a feature where people can tell the system to delete all the data provided so they can exercise their “right to be forgotten”.