DEEP LEARNING IN SPEAKER IDENTIFICATION: MODERN METHODS AND DEVELOPMENT PROSPECTS

Abstract

This scientific article is a review of research on the use of deep learning in the problem of speaker identification. The article will examine the problems of voice identification, highlighting the relevance and need for effective methods in this area. Modern methods for speaker identification will be considered, as well as prospects for the development of such systems.This article has two objectives and these are “Comparative analysis with traditional methods”, “Review of the current state of technology”. “Comparative Analysis with Traditional Methods” highlights the key differences and advantages of deep learning compared to traditional approaches to speaker identification. "State of the Art Review" provides a comprehensive overview of current deep learning methods used for speaker identification.