MINIMUM SPEECH SIGNAL DATA AS SPECIAL IDENTIFICATION INDICATORS IN PHONOSCOPY

Abstract

The physical parameters of the speech signal due to physiological, psychological and other factors, as well as those
associated with voice biometrics, are widely used in the field of forensic examination and phonoscopy. There are different methods for identifying and verifying a person using voice biometric data. As a result of our research, it was revealed that in phonoscopic studies, the minimum data of the speech signal is more reliable and stable. During the experiment, the study also involved the vocal materials of monozygotic twins, which are considered identical in terms of genetic data. Based on the results of the experiment, we came to the conclusion that the “voice prints” in monozygous
twins are of an individual character.
The article presents the results of three experiments carried out to identify the significance of the minimum data of
the speech signal as a marker of individuality. In the course of the experiment, it turned out that in a specific position
for a certain sound, the average temporal indicators are individual in nature and are repeated, since this parameter depends on the respiratory process and the geometric structure of the speech apparatus. The results of the experiments
proved the informativeness and special importance of the minimum data of the speech signal for personal identification