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

2013 , ТОМ 13, НОМЕР 1 ( ЯНВАРЬ-ФЕВРАЛЬ )

ISSN 2226-1494 (print), ISSN 2500-0373 (online)

Меню

Публикации

Главный редактор

НИКИФОРОВ
Владимир Олегович
д.т.н., профессор

Партнеры

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

Аннотация

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.

Ключевые слова:

speech segments clustering, Bayesian variation analysis, speech technologies

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License