Coughing and voice analysis can help emergency services identify critical cases of the lung disease caused by the Sars-CoV-2 virus needing rapid intervention. The Luxembourg Institute of Science and Technology (LIST) is looking to develop artificial intelligence solutions to help speed up diagnosis.
To develop the system, the researchers need as many voice samples as possible to help machine learning. “The more data and completed surveys the system receives, the more it can learn and improve in the battle against covid-19,” LIST said in a press release.
The research team--in cooperation with the Luxemburg Institute of Health and the University of Luxembourg--has set up an online platform to record voice samples in a range of languages, including English.
Voice samples of covid-19 patients and those who have not been diagnosed with the disease are needed for the project.
As part of its work on coronavirus, LIST also carries out weekly analysis of wastewater in Luxembourg to measure virus particles. The latest measurements from last week continued showing high levels of the virus in wastewater, roughly three times the amount measured at the peak of the first wave at the start of April.
“No improvement of the overall situation is observed for the moment,” the report said.
Researchers from the covid-19 task force at the end of last week said the situation had slightly relaxed one week after a nightly curfew and rule of four came into force, with the spread of the virus somewhat slower than before.
“But the high numbers still require continuing social efforts to mitigate the current epidemic wave and bring down the numbers into a better controllable regime,” a report from 6 November said. “The more constant but high number of daily cases is still putting contact tracing at its limits.”