Speaker-adaptive speech recognition using speaker diarization for improved transcription of large spoken archives
Identifikátory výsledku
Kód výsledku v IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F46747885%3A24220%2F13%3A%230002801" target="_blank" >RIV/46747885:24220/13:#0002801 - isvavai.cz</a>
Výsledek na webu
<a href="http://dx.doi.org/10.1016/j.specom.2013.06.017" target="_blank" >http://dx.doi.org/10.1016/j.specom.2013.06.017</a>
DOI - Digital Object Identifier
<a href="http://dx.doi.org/10.1016/j.specom.2013.06.017" target="_blank" >10.1016/j.specom.2013.06.017</a>
Alternativní jazyky
Jazyk výsledku
angličtina
Název v původním jazyce
Speaker-adaptive speech recognition using speaker diarization for improved transcription of large spoken archives
Popis výsledku v původním jazyce
This paper deals with speaker-adaptive speech recognition for large spoken archives. The goal is to improve the recognition accuracy of an automatic speech recognition (ASR) system that is being deployed for transcription of a large archive of Czech radio. This archive represents a significant part of Czech cultural heritage, as it contains recordings covering 90 years of broadcasting. A large portion of these documents (100,000 h) is to be transcribed and made public for browsing. To improve the transcription results, an efficient speaker-adaptive scheme is proposed. The scheme is based on integration of speaker diarization and adaptation methods and is designed to achieve a low Real-Time Factor (RTF) of the entire adaptation process, because the archive's size is enormous. It thus employs just two decoding passes, where the first one is carried out using the lexicon with a reduced number of items. Moreover, the transcripts from the first pass serve not only for adaptation, but also a
Název v anglickém jazyce
Speaker-adaptive speech recognition using speaker diarization for improved transcription of large spoken archives
Popis výsledku anglicky
This paper deals with speaker-adaptive speech recognition for large spoken archives. The goal is to improve the recognition accuracy of an automatic speech recognition (ASR) system that is being deployed for transcription of a large archive of Czech radio. This archive represents a significant part of Czech cultural heritage, as it contains recordings covering 90 years of broadcasting. A large portion of these documents (100,000 h) is to be transcribed and made public for browsing. To improve the transcription results, an efficient speaker-adaptive scheme is proposed. The scheme is based on integration of speaker diarization and adaptation methods and is designed to achieve a low Real-Time Factor (RTF) of the entire adaptation process, because the archive's size is enormous. It thus employs just two decoding passes, where the first one is carried out using the lexicon with a reduced number of items. Moreover, the transcripts from the first pass serve not only for adaptation, but also a
Klasifikace
Druh
J<sub>x</sub> - Nezařazeno - Článek v odborném periodiku (Jimp, Jsc a Jost)
CEP obor
JC - Počítačový hardware a software
OECD FORD obor
—
Návaznosti výsledku
Projekt
<a href="/cs/project/GPP103%2F11%2FP499" target="_blank" >GPP103/11/P499: Metody adaptace na mluvčího v systémech rozpoznávání řeči</a><br>
Návaznosti
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)<br>S - Specificky vyzkum na vysokych skolach
Ostatní
Rok uplatnění
2013
Kód důvěrnosti údajů
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Údaje specifické pro druh výsledku
Název periodika
SPEECH COMMUNICATION
ISSN
0167-6393
e-ISSN
—
Svazek periodika
55
Číslo periodika v rámci svazku
10
Stát vydavatele periodika
NL - Nizozemsko
Počet stran výsledku
14
Strana od-do
1033-1046
Kód UT WoS článku
000324227500009
EID výsledku v databázi Scopus
—