Publication Details

BUT QUESST 2015 System Description

SKÁCEL Miroslav and SZŐKE Igor. BUT QUESST 2015 System Description. In: CEUR Workshop Proceedings. Wurzen: CEUR-WS.org, 2015, pp. 1-3. ISSN 1613-0073.
Czech title
BUT QUESST 2015 Popis systému
Type
conference paper
Language
english
Authors
Skácel Miroslav, Ing. (DCGM FIT BUT)
Szőke Igor, Ing., Ph.D. (DCGM FIT BUT)
URL
Keywords

query-by-example, DTW, dynamic time warping, spoken term detection, keyword spotting, bottle-neck features

Abstract

All our systems are based on Dynamic Time Warping (DTW). These systems use bottle-neck features (BN) as input. The bottle-neck feature extractors were trained on Global Phone Czech, Portuguese, Russian and Spanish languages, so our approach is in low-resource category. We also aimed onT1/T2/T3 types of query search for late submission sys-tems. System calibration and fusion were based on binarylogistic regression.

Annotation

This paper describes system for query-by-example spoken term detection built for QUESST 2015 evaluation. The system uses dynamic time warping to detect spoken term utilizing bottle-neck features.

Published
2015
Pages
1-3
Journal
CEUR Workshop Proceedings, vol. 2015, no. 1436, ISSN 1613-0073
Proceedings
CEUR Workshop Proceedings
Conference
MediaEval 2015, Wurzen, DE
Publisher
CEUR-WS.org
Place
Wurzen, DE
EID Scopus
BibTeX
@INPROCEEDINGS{FITPUB11069,
   author = "Miroslav Sk\'{a}cel and Igor Sz\H{o}ke",
   title = "BUT QUESST 2015 System Description",
   pages = "1--3",
   booktitle = "CEUR Workshop Proceedings",
   journal = "CEUR Workshop Proceedings",
   volume = 2015,
   number = 1436,
   year = 2015,
   location = "Wurzen, DE",
   publisher = "CEUR-WS.org",
   ISSN = "1613-0073",
   language = "english",
   url = "https://www.fit.vut.cz/research/publication/11069"
}
Back to top