Publication Details
Spoken Term Detection System Based on a Combination of LVCSR and Phonetic Search
Fapšo Michal, Ing., Ph.D.
Karafiát Martin, Ing., Ph.D. (DCGM)
Burget Lukáš, doc. Ing., Ph.D. (DCGM)
Grézl František, Ing., Ph.D. (DCGM)
Schwarz Petr, Ing., Ph.D. (DCGM)
Glembek Ondřej, Ing., Ph.D.
Matějka Pavel, Ing., Ph.D. (DCGM)
Kopecký Jiří, Ing.
Černocký Jan, prof. Dr. Ing. (DCGM)
spoken term detection
The paper describes Spoken Term Detection System Based on a Combination of LVCSR and Phonetic Search.
The paper presents Brno University of Technology (BUT) system for indexing and search of speech, combining LVCSR and phonetic approach. It brings a complete description of individual building blocks of the system from signal processing, through the recognizers, indexing and search until the normalization of detection scores. It also describes the data used in the first edition of NIST Spoken term detection (STD) evaluation. The results are presented on three US-English conditions - meetings, broadcast news and conversational telephone speech, in terms of detection error trade-off (DET) curves and term-weighted values (TWV) metrics defined by NIST.
@misc{BUT63690,
author="Igor {Szőke} and Michal {Fapšo} and Martin {Karafiát} and Lukáš {Burget} and František {Grézl} and Petr {Schwarz} and Ondřej {Glembek} and Pavel {Matějka} and Jiří {Kopecký} and Jan {Černocký}",
title="Spoken Term Detection System Based on a Combination of LVCSR and Phonetic Search",
year="2007",
pages="1--1",
address="Brno",
url="http://www.fit.vutbr.cz/research/groups/speech/publi/2007/szoke_kws_ins_mlmi_poster_2007.pdf",
note="presentation, poster"
}