Автоматизована система ідентифікації особи за її голосом

Abstract

The work is devoted to the development of methods, algorithms and software for the automatic identification of the speaker by his voice in systems of authorized access to information resources in security systems. In the work, the mathematical foundations were considered and algorithms and programs for the analysis of speech signals were developed, algorithms and programs were developed for describing the speaker by features such as the frequency of the main tone and the short-term spectrum, and a neural network algorithm for the classification of the announcer based on his feature description was developed.