Skip to content

Detail of publication


Pražák, A. and Müller, L. and Psutka Josef V. : LVCSR system for automatic online subtitling . SPECOM 2005 proceedings, p. 325-328, Moscow State Linguistics University, Moscow, 2005.


In this paper, a LVCSR system for automatic online subtitling (closed captioning) of the tasks with a vocabulary limited by a specific domain is proposed. The recognition system uses lexical trees and a bigram language model. The first part of this article is focused on recognition network creation and its decoding with respect to real-time operation demands. The second part outlines the application framework allowing subtitle generation and displaying for any audio/video source. Finally, experimental results of weather forecast and TV ice hockey commentary are reported.

Detail of publication

Title: LVCSR system for automatic online subtitling
Author: Pražák, A. ; Müller, L. ; Psutka Josef V.
Language: English
Date of publication: 17 Oct 2005
Year: 2005
Type of publication: Papers in proceedings of reviewed conferences
Title of journal or book: SPECOM 2005 proceedings
Page: 325 - 328
ISBN: 5-7452-0110-X
Publisher: Moscow State Linguistics University
Address: Moscow
Date: 17 Oct 2005 - 19 Oct 2005
/ 2011-06-09 13:06:51 /


speech recognition, LVCSR, lexical tree, bigram language model, real-time, DirectX


 author = {Pra\v{z}\'{a}k, A. and M\"{u}ller, L. and Psutka Josef V.},
 title = {LVCSR system for automatic online subtitling},
 year = {2005},
 publisher = {Moscow State Linguistics University},
 journal = {SPECOM 2005 proceedings},
 address = {Moscow},
 pages = {325-328},
 ISBN = {5-7452-0110-X},
 url = {},