Voice Changes as a Result of Psychoemotional Impact

The study object of this work is the voice-speech signal as a presumed source of objective physical parameters for identifying the psychoemotional state. The purpose of the work is to search for objective characteristics of a voice-speech signal, indicating a particular psychoemotional state. In the process of work, an acoustic and correlation analysis of people's voices was carried out from an open database, which contains audio recordings of a voice-speech signal for 5 basic psycho-emotional states: happiness, sadness, anger, fear and a neutral state. As a result of the study, the averaged acoustic parameters (jitter, shimmer, harmonics-to-noise-ratio, spectral characteristics) were calculated for each experimental group, averaged behavior models of the fundamental frequency for various emotions were obtained, the relationship between changes in the intensity of the voice-speech signal and emotions was refuted, the relationship between the spectral characteristics of the voice-speech signal and a certain emotional state was found. Thus, objective voice-speech “biomarkers” of the psychoemotional state were identified.