DC Field | Value | Language |
---|---|---|
dc.contributor.author | Popović, Branislav | en |
dc.contributor.author | Janev, Marko | en |
dc.contributor.author | Delić, Vlado | en |
dc.date.accessioned | 2020-04-27T10:55:17Z | - |
dc.date.available | 2020-04-27T10:55:17Z | - |
dc.date.issued | 2012-12-01 | en |
dc.identifier.isbn | 978-1-467-32984-2 | en |
dc.identifier.uri | http://researchrepository.mi.sanu.ac.rs/handle/123456789/899 | - |
dc.description.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. | en |
dc.publisher | IEEE | - |
dc.relation.ispartof | 2012 20th Telecommunications Forum, TELFOR 2012 - Proceedings | en |
dc.subject | continuous speech recognition | Gaussian selection | hierarchical clustering | split-and-merge | en |
dc.title | Gaussian selection algorithm in Continuous Speech Recognition | en |
dc.type | Conference Paper | en |
dc.relation.conference | 20th Telecommunications Forum, TELFOR 2012; Belgrade; Serbia; 20 November 2012 through 22 November 2012 | - |
dc.identifier.doi | 10.1109/TELFOR.2012.6419307 | en |
dc.identifier.scopus | 2-s2.0-84874178314 | en |
dc.contributor.affiliation | Mathematical Institute of the Serbian Academy of Sciences and Arts | - |
dc.relation.firstpage | 705 | en |
dc.relation.lastpage | 712 | en |
item.grantfulltext | none | - |
item.cerifentitytype | Publications | - |
item.fulltext | No Fulltext | - |
item.openairecristype | http://purl.org/coar/resource_type/c_18cf | - |
item.openairetype | Conference Paper | - |
crisitem.author.orcid | 0000-0003-3246-4988 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.