Voice based detection of type 2 diabetes mellitus

There has been much recent research using voice parameters of pathological persons. In this paper, research has been carried out to distinguish the vocal characteristics of patients having type 2 diabetes mellitus with healthy persons. The study has been carried out on 177 voice samples in age group of 40-60 years including healthy and diabetic males and females. The acoustic analysis was done using MDVP. The voice parameters like jitter, shimmer, smoothed amplitude perturbation quotient, noise to harmonic ratio, relative average perturbation, amplitude perturbation quotient show significant difference in their values for type 2 diabetes mellitus patients as compared to non-diabetic persons. We see the results as an important step towards non- invasive diagnosis of type 2 DM disease.