AI can detect diabetes from 10 sec of voice, study finds

4 Min Read

Researchers from Klick Labs have made significant strides in diabetes detection by combining voice technology with artificial intelligence (AI).

Their groundbreaking study, recently published in Mayo Clinic Proceedings: Digital Health, showcases the potential of using voice analysis, along with basic health data, to accurately identify type 2 diabetes in individuals.

The AI model developed in the study demonstrates impressive accuracy rates, with 89% for women and 86% for men.

Voice Analysis for Diabetes Detection

The study involved 267 participants, both non-diabetic and type 2 diabetic individuals, who were asked to record a specific phrase on their smartphones multiple times a day for two weeks.

From over 18,000 voice recordings, researchers examined 14 acoustic features to identify distinctions between non-diabetic and type 2 diabetic individuals.

These vocal variations, which are often imperceptible to the human ear, were analyzed using signal processing techniques.

Gender-Specific Vocal Changes

Remarkably, the study revealed that the vocal changes associated with type 2 diabetes manifested differently in males and females.

Changes in pitch and intensity were among the vocal features analyzed. These findings offer a potential avenue for developing gender-specific diagnostic tools.

Addressing a Global Health Challenge

Type 2 diabetes is a widespread global health issue, affecting nearly 90% of diabetic cases. Alarmingly, almost half of the 240 million adults living with diabetes worldwide are unaware of their condition.

Current diagnostic tests for diabetes typically require a visit to a healthcare provider and may not be accessible or convenient for everyone.

A Non-Intrusive Screening Tool

The voice analysis technology developed by Klick Labs has the potential to revolutionize diabetes screening by providing a non-intrusive and accessible method.

This approach could help identify a significant percentage of undiagnosed individuals with type 2 diabetes. By eliminating barriers related to time, travel, and cost, voice technology offers a promising solution to this healthcare challenge.

The Path Forward

Yan Fossat, vice president of Klick Labs, emphasized that this breakthrough could extend to other areas, such as prediabetes, women’s health, and hypertension.

The research underscores the vast potential of voice technology in healthcare as an affordable and accessible digital screening tool.

Future studies will explore these possibilities and expand the application of voice technology in healthcare diagnostics.


Klick Labs’ pioneering research in voice technology and AI has the potential to transform the early detection of type 2 diabetes and other health conditions.

By harnessing the power of voice analysis, this innovative approach could help millions of individuals worldwide receive timely diagnoses and access essential healthcare interventions.

The study’s success highlights the promising role of technology in improving healthcare practices and enhancing accessibility to diagnostic tools.

If you care about diabetes, please read studies about Plant-based protein foods may help reverse type 2 diabetes and honey could help control blood sugar.

For more information about diabetes, please see recent studies about Vitamin D that may reduce dangerous complications in diabetes and results showing plant-based protein foods may help reverse type 2 diabetes.

Share This Article
By admin
test bio
Leave a comment