Classifying and Standardizing Parameters of Voice towards Categorizing Individual Variations for Personalised Biomedicine

Abstract

Most of the eminent scholars of Unani medicine like Hippocrates, Galen and Avicenna have mentioned about four temperaments on the basis of four types of Humours. They believed that every person has certain characteristic features morphologically, physiologically and psychologically, which are distinct from one another on the basis of dominant humour. The normal proportion of these humours with each other is responsible for health while their imbalance causes disease and it has diagnostic and treatment values. Voice is one of the physiological functions and has been extensively documented in the literature of traditional medicine. Therefore, an adequate dataset of actual voice samples was taken from the random population and it was hypothesized that categorization of people can be done on the bases of their voice characteristics. There is a strong co-relation between voice characteristics and temperament of different individuals, research on voice characteristics was taken to interpret the observations in terms of four temperaments which was not done earlier. The study was carried out on healthy volunteers of the age group between 25–35 years (both the sexes). Zoom recorder was used to record the voice samples and then with the help of PRAAT software a voice report was obtained (shows readings of various voice parameters). Among various parameters in the voice report only eight were chosen according to the need of the study. Then statistical analysis was done to find the differences of individuals on the basis of their voice characteristics. Data suggests that not all voice parameters are equally efficient for categorization; however, rate of speech, pulse, and voice breaks can be used to divide people into four broad categories. The clinical advantage of categorizing voice is discussed with an aim to develop it as a tool for personalized medicine.