|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
checked on Aug 14, 2022
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.