The use of voice data for health purposes is a vast field of investigation that has so far been little explored. With regard to diabetes in particular, voice data is a promising data source for identifying, based on various acoustic characteristics, digital voice biomarkers associated with stress, anxiety, fear of hypoglycaemia, emotions and clinical outcomes such as cardiovascular disease or blood sugar control.
The most common phonatory symptoms in type 2 diabetes are vocal fatigue, tiredness and hoarseness. They are more common in people with diabetes than in the general population. It appears possible to identify, in a non-invasive manner, useful voice biomarkers for the management of diabetes and the prevention of diabetes-related complications.
This project aims to identify voice biomarkers for designing just-in-time intervention studies. This project is dedicated to the identification of voice biomarkers associated with patient emotions.
Guy FAGHERAZZI, PhD
1 AB rue Thomas Edison