Title: Refinement approach for adaptation based on combination of MAP and fMLLR
Other Titles: Zlepšený přístup k adaptaci založené na kombinaci MAP a fMLLR
Authors: Zají­c, Zbyněk
Machlica, Lukáš
Müller, Luděk
Citation: ZAJÍC, Zbyněk; MACHLICA, Lukᚡ; MÜLLER, Luděk. Refinement approach for adaptation based on combination of MAP and fMLLR. In: Text, speech and dialogue. Berlin: Springer, 2009, p. 274-281. (Lecture notes in computer science; 5729). ISBN 978-3-642-04207-2.
Issue Date: 2009
Publisher: Springer
Document type: článek
URI: http://www.kky.zcu.cz/cs/publications/ZbynekZajic_2009_RefinementApproach
ISBN: 978-3-642-04207-2
Keywords: adaptace;fMLLR;MAP
Keywords in different language: adaptation;fMLLR;MAP
Abstract in different language: This paper deals with a combination of basic adaptation techniques of Hidden Markov Model used in the speech recognition. The adaptation methods approach the data only through their statistics, which have to be accumulated before the adaptation process. When performing two adaptations subsequently, the data statistics have to be accumulated twice in each of the adaptation passes. However, when the adaptation methods are chosen with care, the data statistics may be accumulated only once, as proposed in this paper. This significantly reduces the time consumption and avoids the need to store all the adaptation data. Combination of Maximum A-Posteriori Probability and feature Maximum Likelihood Linear Regression adaptation is considered. Motivation for such an approach could be the on-line adaptation, where the time consumption is of big importance.
Rights: © Zbyněk Zají­c - Lukáš Machlica - Luděk Müller
Appears in Collections:Články / Articles (KKY)
Články / Articles (NTIS)

Files in This Item:
File Description SizeFormat 
ZbynekZajic_2009_RefinementApproach.pdfPlný text191,73 kBAdobe PDFView/Open

Please use this identifier to cite or link to this item: http://hdl.handle.net/11025/16954

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