Authors: Popović, Branislav
Janev, Marko 
Delić, Vlado
Affiliations: Mathematical Institute of the Serbian Academy of Sciences and Arts 
Title: Gaussian selection algorithm in Continuous Speech Recognition
Journal: 2012 20th Telecommunications Forum, TELFOR 2012 - Proceedings
First page: 705
Last page: 712
Conference: 20th Telecommunications Forum, TELFOR 2012; Belgrade; Serbia; 20 November 2012 through 22 November 2012
Issue Date: 1-Dec-2012
ISBN: 978-1-467-32984-2
DOI: 10.1109/TELFOR.2012.6419307
Abstract: 
Clustering of Gaussian mixture components, i.e. Hierarchical Gaussian mixture model clustering (HGMMC) is a key component of Gaussian selection (GS) algorithm, used in order to increase the speed of a Continuous Speech Recognition (CSR) system, without any significant degradation of its recognition accuracy. In this paper a novel Split-and-Merge (S&M) HGMMC algorithm is applied to GS, in order to achieve a better trade-off between speed and accuracy in a CSR task. The algorithm is further improved by introducing model selection in order to obtain the best possible trade-off between recognition accuracy and computational load in a GS task applied within an actual recognition system. At the end of the paper we discuss additional improvements towards finding the optimal setting for the Gaussian selection scheme.
Keywords: continuous speech recognition | Gaussian selection | hierarchical clustering | split-and-merge
Publisher: IEEE

Show full item record

Page view(s)

15
checked on Aug 14, 2022

Google ScholarTM

Check

Altmetric

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.