. . . "Automatic Online Subtitling of the Czech Parliament Meetings" . "6"^^ . "Lecture Notes in Artificial Intelligence" . "8"^^ . "Automatick\u00E9 online titulkov\u00E1n\u00ED parlamentn\u00EDch p\u0159enos\u016F"@cs . . "This paper describes a LVCSR system for automatic online subtitling (closed captioning) of TV transmissions of the Czech Parliament meetings. The recognition system is based on Hidden Markov Models, lexical trees and bigram language model. The acoustic model is trained on 40 hours of parliament speech and the language model on more than 10M tokens of parliament speech trancriptions. The first part of the article is focused on text normalization and class-based language model preparation. The second part describes the recognition network and its decoding with respect to real-time operation demands using up to 100k vocabulary. The third part outlines the application framework allowing generation and displaying of subtitles for any audio/video source. Finally, experimental results obtained on parliament speeches with recognition accuracy varying from 80 to 95 % (according to the discussed topic) are reported and discussed."@en . "[095D44D867DE]" . "501" . "6"^^ . . "Hoidekr, Jan" . . "Automatick\u00E9 online titulkov\u00E1n\u00ED parlamentn\u00EDch p\u0159enos\u016F"@cs . "DE - Spolkov\u00E1 republika N\u011Bmecko" . "Automatic Online Subtitling of the Czech Parliament Meetings"@en . . . "Automatic Online Subtitling of the Czech Parliament Meetings" . . "ASR; online; subtitling; parliament; czech"@en . "Psutka, Josef" . . . "Rozpozn\u00E1vac\u00ED syst\u00E9m je zalo\u017Een na skryt\u00FDch markovov\u00FDch modelech (HMM), lexik\u00E1ln\u00EDch stromech a bigramov\u00E9m jazykov\u00E9m modelu. Akustick\u00FD model je natr\u00E9nov\u00E1n na 40 hodin\u00E1ch parlamentn\u00EDch sch\u016Fz\u00ED a jazykov\u00FD model na v\u00EDce ne\u017E 10M slov p\u0159epis\u016F parlamentn\u00EDch sch\u016Fz\u00ED. Prvn\u00ED \u010D\u00E1st \u010Dl\u00E1nku se zab\u00FDv\u00E1 normalizac\u00ED textu a p\u0159\u00EDpravou t\u0159\u00EDdov\u00E9ho jazykov\u00E9ho modelu. Druh\u00E1 \u010D\u00E1st popisuje rozpozn\u00E1vac\u00ED s\u00ED\u0165 a jej\u00ED dek\u00F3dov\u00E1n\u00ED s ohledem pr\u00E1ci v re\u00E1ln\u00E9m \u010Dase se slovn\u00EDkem a\u017E 100k slov. T\u0159et\u00ED \u010D\u00E1st nasti\u0148uje strukturu aplikace umo\u017E\u0148uj\u00EDc\u00ED generov\u00E1n"@cs . "P(1QS101470516)" . . "RIV/49777513:23520/06:00000013" . "0" . . . . "466411" . . "Automatic Online Subtitling of the Czech Parliament Meetings"@en . "M\u00FCller, Lud\u011Bk" . . . . "Pra\u017E\u00E1k, Ale\u0161" . . "0302-9743" . "Kanis, Jakub" . . "RIV/49777513:23520/06:00000013!RIV07-AV0-23520___" . "23520" . . "This paper describes a LVCSR system for automatic online subtitling (closed captioning) of TV transmissions of the Czech Parliament meetings. The recognition system is based on Hidden Markov Models, lexical trees and bigram language model. The acoustic model is trained on 40 hours of parliament speech and the language model on more than 10M tokens of parliament speech trancriptions. The first part of the article is focused on text normalization and class-based language model preparation. The second part describes the recognition network and its decoding with respect to real-time operation demands using up to 100k vocabulary. The third part outlines the application framework allowing generation and displaying of subtitles for any audio/video source. Finally, experimental results obtained on parliament speeches with recognition accuracy varying from 80 to 95 % (according to the discussed topic) are reported and discussed." .