20 seconds of speech could help detect type 2 diabetes using AI-based tool

An AI speech model developed by deep tech company thymia with RMIT University gave a higher type 2 diabetes risk score to people who reported having the condition than to those who did not 80% of the time, according to research presented at the European Association for the Study of Diabetes (EASD) annual meeting in Milan, Italy, Sept. 28–Oct. 2. The model, trained on 63,283 voice samples from 21,129 people in the U.K. and U.S. and validated on 20-second recordings of people reading an Aesop's fable, achieved 82% sensitivity and a 47% false-positive rate in a subgroup of 801 participants who took at-home HbA1c tests within three months of the recording. The study's authors say the tool could offer a fast, noninvasive and scalable screening method, noting that about 30% of U.K. type 2 diabetes cases are undiagnosed and fewer than half of eligible adults (40.4%) attend NHS diabetes screening appointments.

Just 20 seconds of speech could help detect type 2 diabetes using AI-based tool Sadie Harley Scientific Editor Andrew Zinin Chief Editor Signs of type 2 diabetes can be detected in just a few seconds from the way someone talks, according to research being presented at the annual meeting of the European Association for the Study of Diabetes EASD https://www.easd.org/annual-meeting/easd-2026/ in Milan, Italy, Sept. 28–Oct. 2. The study, the largest of its kind, concludes that AI-based analysis of short recordings could be a fast, noninvasive and scalable screening tool for the condition. Type 2 diabetes is becoming increasingly common, and early detection is key to preventing complications such as heart disease and nerve damage. Yet many cases go undetected, putting pressure on health care systems. In the U.K., for example, about 30% of cases are undiagnosed, and around 60% of the £10.7 billion the National Health Service NHS spends on diabetes each year goes toward managing complications. Screening currently involves blood tests or appointments with primary care doctors. For example, the NHS includes diabetes screening in the health checks it offers to people 40 and older every five years. But fewer than half of eligible adults 40.4% attend these useful but time-consuming appointments, the study's authors say. Previous research has linked type 2 diabetes to changes in speech, such as increased hoarseness and roughness and reduced control of breath and voice while speaking. Researchers at deep tech company thymia, led by senior machine learning researcher Roseline Polle and senior research scientist Dr. Elisa Brann, worked with colleagues at RMIT University in Melbourne, Australia, to develop an AI model to detect these changes. They trained it on 63,283 voice samples from 21,129 people in the U.K. and U.S. who had reported whether they had been diagnosed with diabetes. They then validated, or tested, the speech model on remote 20-second recordings of people reading one of Aesop's fables short stories with animal characters and moral lessons . The first of two evaluations compared the model's performance on recordings made by 7,319 adults in the U.K. 67% female; 45.8% age 40 or older; 217 reported having type 2 diabetes . The speech model gave a higher risk score to people who reported having type 2 diabetes than to those who did not report having the condition 80% of the time, a result considered clinically useful. The model performed well across different sexes and ages. However, performance was lower on recordings from Black participants. This was likely due to the low number of Black participants reporting type 2 diabetes, the authors say. Performance was also lower on recordings from people with heart disease, high blood pressure or obesity. This is thought to be because these conditions often coincide with type 2 diabetes and may cause similar vocal changes. The second evaluation involved a subgroup of 801 participants who took HbA1c tests https://medicalxpress.com/news/2022-11-screening-older-adults-aged-.html?utm source=embeddings&utm medium=related&utm campaign=internal at home within three months of the speech recording. The HbA1c blood test measures average blood sugar levels over the past two to three months and is the gold-standard test for type 2 diabetes. It can also detect prediabetes, in which blood sugar levels are higher than normal but not high enough to be classified as diabetes. In this evaluation, the speech model gave a higher risk score to people with type 2 diabetes than to those who did not have the condition, based on HbA1c tests, 75% of the time. The sensitivity was 82% that is, the model correctly identified 82% of the people with type 2 diabetes , and the false-positive rate was 47%. The speech model was also able to distinguish among people at low, medium and high risk of type 2 diabetes, based on their HbA1c results. None of the people the model classified as low risk had blood results in the diabetic or prediabetic range. The researchers conclude that type 2 diabetes can be detected at clinically useful levels from 20 seconds of speech. Pending further clinical validation of the tool, one option would be for primary care doctors to use short recordings to triage patients. Those at higher risk could then be given blood tests to confirm their status. Speech-based screening would sit alongside blood testing, not replace it, the authors say. Giedrė Čepukaitytė, a research scientist at thymia who will present the findings at EASD, says, "This is the largest real-world study of speech-based screening for type 2 diabetes to date that also checks the model's predictions against blood test results as well as against what people reported about their own diagnosis. Those flagged as higher risk by the model had blood results to match. "This has the potential to change what screening looks like. A speech sample can be taken over the phone or through an app, so we can reach far more of the people who need a blood test than current approaches do, particularly those who never get to a health check. "Our model opens a new route to screening for diabetes. It is not a replacement for a blood test, and it should never stop anyone who thinks they need one from getting one. "Our next step is to test the model in clinical settings and to understand how well it works for every group of people, because a screening tool has to work for everyone." Key medical concepts Diabetes Type 2 Citation : Just 20 seconds of speech could help detect type 2 diabetes using AI-based tool 2026, September 28 retrieved 29 September 2026 from https://medicalxpress.com/news/2026-09-seconds-speech-diabetes-ai-based.html