Skip to content

Publications

Detail of publication

Citation

Jurčíček, F. and Pražák, A. and Müller, L. and Psutka, J. and Šmídl, L. : Design of LVCSR decoder for Czech language . ECMS 2003, p. 39-43, Technical University , Liberec , 2003.

Abstract

The LVCSR systems are developed for several decades. Nowadays LVCSR system take advantage of increasing performance of computers and sophisticated algorithms. There is permanent effort to integrate larger lexicons into LVCSR system. A n-gram language model can improve recognition accuracy. However, memory requirments are substantial, so in practise for real-time implementation of LVCSR system only a bigram language model is often considered. The LVCSR system can be used for either automatic dictation or as speech recognition module (for voice dialog system) loosely coupled with a speech understanding module. The output of the LVCSR system is either a word lattice, or the best sequence of words matching the input acustic signal. The word lattice can be processed by higher level blocks.

Detail of publication

Title: Design of LVCSR decoder for Czech language
Author: Jurčíček, F. ; Pražák, A. ; Müller, L. ; Psutka, J. ; Šmídl, L.
Language: English
Date of publication: 2 Jun 2003
Year: 2003
Type of publication: Papers in proceedings of reviewed conferences
Title of journal or book: ECMS 2003
Page: 39 - 43
ISBN: 807083708X
Publisher: Technical University
Address: Liberec
Date: 2 Jun 2003 - 4 Jun 2003
/ /

Keywords

speech recognition, real-time, language models

BibTeX

@INPROCEEDINGS{JurcicekF_2003_DesignofLVCSR,
 author = {Jur\v{c}\'{i}\v{c}ek, F. and Pra\v{z}\'{a}k, A. and M\"{u}ller, L. and Psutka, J. and \v{S}m\'{i}dl, L.},
 title = {Design of LVCSR decoder for Czech language},
 year = {2003},
 publisher = {Technical University },
 journal = {ECMS 2003},
 address = {Liberec },
 pages = {39-43},
 ISBN = {807083708X     },
 url = {http://www.kky.zcu.cz/en/publications/JurcicekF_2003_DesignofLVCSR},
}