AGGLOMERATIVE CLUSTERING OF SOUND RECORD SPEECH SEGMENTS BASED ON BAYESIAN INFORMATION CRITERION

O. Y. Kydashev


Read the full article 

Abstract

This paper presents the detailed description of agglomerative clustering system implementation for speech segments based on Bayesian information criterion. Numerical experiment results with different acoustic features, as well as the full and diagonal covariance matrices application are given. The error rate DER equal to 6.4% for audio records of radio «Svoboda» was achieved by means of designed system.


Keywords: speech segments clustering, Bayesian variation analysis, speech technologies
Copyright 2001-2017 ©
Scientific and Technical Journal
of Information Technologies, Mechanics and Optics.
All rights reserved.

Яндекс.Метрика