The acoustic-phonetic-based speaker recognition is one of the fastest growing scientific areas, but the problem is still not solved. The subject of the candidate has the opportunity to develop new methods, new acoustic-phonetic features for speaker recognition and put them into practice.