Ten seconds of speech can flag if someone has diabetes

Just 10 seconds of recorded speech is needed to tell if someone has diabetes, new scientific developments have shown.

(That's good news for those scared of needles).

A voice recording, along with basic health data such as age, sex, height and weight is all that an artificial intelligence (AI) program needs to establish whether someone is diabetic.

The system has nearly 90 per cent accuracy, too.

As part of the peer-reviewed study published in the Mayo Clinic Proceedings journal, Klick Labs got 267 people who had recently completed a standard test for Type 2 diabetes.

Each of them recorded a short phrase on their phones six times a day for two weeks and researchers with the help of AI analysed over 18,000 samples to determine the acoustic differences such as pitch between people who are diabetic and those who are not.

“Voice synthesis is a complex process that relies on the combined effects of the respiratory system, the nervous system, and the larynx," the researchers wrote.

"Anything that affects these systems can influence the voice," and there are some changes that the human ear wouldn't be able to pick up while the AI program can.

The results were 86 per cent accurate for men, and 89 per cent for women.

At the moment, a GP visit along with urine and blood tests are required for a diabetes test but this technology could change that.

Jaycee Kaufman, author of the study explained how this could "transform" the diabetes diagnosis process.

“Our research highlights significant vocal variations between individuals with and without Type 2 diabetes and could transform how the medical community screens for diabetes,” she said in a statement.

“Current methods of detection can require a lot of time, travel, and cost. Voice technology has the potential to remove these barriers entirely.”

“Voice technology could revolutionize healthcare practices as an accessible and affordable digital screening tool,” said co-author Yan Fossat added.

It's not just diabetes as voice recordings can also be used to diagnose other illnesses such as Covid, according to past research.

Sign up to our free Indy100 weekly newsletter

Have your say in our news democracy. Click the upvote icon at the top of the page to help raise this article through the indy100 rankings.