An app has been developed that can detect Covid-19 in your voice – and we’re hoping it’ll be here before we all start testing more regularly in winter.
Scientists say the AI-powered technology is easier to use and is more precise than a lateral flow test. It takes less than a minute to find positive cases and gives the correct result 89% of the time, whilst negative cases are right 83% of the time.
In comparison, the accuracy of nose and throat swabs can vary by brand and drops in people with Covid without symptoms.
The new app could be used to detect cases of Covid before big events such as concerts and sports events. Additionally, it could be used in poorer countries where PCR tests are pricy to buy and hard to distribute.
The Dutch researchers used data from the University of Cambridge’s crowdsourcing Covid-19 Sounds App, which contains 893 audio samples from 4,352 participants, 308 people had tested positive for the virus.
The app is downloaded on to the user’s phone and those involved in the study gave basic details such as demographics, medical history and smoking status. They were then asked to record some respiratory sounds which include coughing three times, breathing deeply through their mouth around three to five times, and reading a short sentence on the screen three times.
A voice technique called Mel-spectrogram analysis was used by the researchers to identify different voice features like loudness, power and variation over time. To differentiate the voices of those with Covid-19 from those who did not have the infection, researchers created different artificial intelligence models and evaluated which one worked best at classifying positive cases.
“These promising results suggest that simple voice recordings and fine-tuned AI algorithms can potentially achieve high precision in determining which patients have Covid-19 infection,” Wafaa Aljbawi, a researcher from the University of Maastricht, said.
“These results show a significant improvement in the accuracy of diagnosing Covid-19 compared to state-of-the-art tests such as the lateral flow test.”
The app is not yet available for public use as the team say more research with more people should be done before the app can begin appearing on people’s phones.
An enormous 53,449 audio samples from 36,116 participants have now been collected and can be used to improve and validate the accuracy of the model since the launch of the project.
The scientists aren’t the only ones looking into this type of technology. In Australia, researchers are working on a separate app to detect Covid-19 if you cough into your phone.