Přístupnostní navigace
E-přihláška
Vyhledávání Vyhledat Zavřít
Detail publikačního výsledku
SZŐKE, I.
Originální název
Keyword Spotting in Meeting Data
Anglický název
Druh
Stať ve sborníku mimo WoS a Scopus
Originální abstrakt
This paper describes several approaches to keyword spotting (KWS) for informal continuous speech. We compare acoustic keyword spotting, spotting in word lattices generated by large vocabulary continuous speech recognition and a hybrid approach making use of phoneme lattices generated by a phoneme recognizer. The systems are compared on carefully defined test data extracted from ICSI meeting database. The acoustic and phoneme-lattice based KWS are based on a phoneme recognizer making use of long temporal split context feature extractionand posterior estimation using neural nets. The advantages and drawbacks of different approaches are discussed.
Anglický abstrakt
Klíčová slova
keyword spotting, lattice, LVCSR, phoneme recognizer, informal speech
Klíčová slova v angličtině
Autoři
Vydáno
27.04.2006
Nakladatel
Faculty of Electrical Engineering and Communication BUT
Místo
Brno
ISBN
80-214-3163-6
Kniha
Proceedings of the 12th Conference Student EEICT 2006 Volume 4
Strany od
440
Strany do
444
Strany počet
5
URL
www.fit.vutbr.cz/~szoke/paper/eeict_2006.pdf
BibTex
@inproceedings{BUT22366, author="Igor {Szőke}", title="Keyword Spotting in Meeting Data", booktitle="Proceedings of the 12th Conference Student EEICT 2006 Volume 4", year="2006", pages="440--444", publisher="Faculty of Electrical Engineering and Communication BUT", address="Brno", isbn="80-214-3163-6", url="www.fit.vutbr.cz/~szoke/paper/eeict_2006.pdf" }