Hypoglycemia is one of the most common and most dangerous acute complications of diabetes. It can lead to dizziness, confusion, unconsciousness or even life-threatening situations within a few minutes. Despite modern glucose sensors, it is often difficult to recognize impending hypoglycaemia in time. At the same time, the human voice is a sensitive mirror of the body: it changes when we are tired, stressed or ill; and, as is now apparent, even when blood sugar levels fall.

Researchers at Inselspital, Bern University Hospital and the University of Bern, together with international partners, have shown for the first time that hypoglycaemia can be reliably detected on the basis of characteristic changes in the voice. Voice recordings made using a standard smartphone microphone, which were then analyzed using a machine learning algorithm, were sufficient for this purpose. Given the widespread use of smartphones, this approach could improve the detection and prevention of hypoglycaemia worldwide, especially in regions where modern glucose sensors are only available to a limited extent.