Copingwith channel mismatch in Query-by-Example - But QUESST 2014

Author

Szoke, Igor ; Skacel, Miroslav ; Burget, Lukas ; Cernocky, Jan

Author_Institution

BUT Speech@FIT, Brno Univ. of Technol., Brno, Czech Republic

fYear

2015

fDate

19-24 April 2015

Firstpage

5838

Lastpage

5842

Abstract

The paper investigates into Query by Example (QbE) - a spoken term detection technique with queries entered by voice. It describes BUT QbE system that achieved the best accuracy in MediaEval QUESST2014 evaluations. This evaluation was challenging because of severe mismatch between queries and utterances, and introduction of new types of queries. The paper provides an analysis of DTW sub-system´s in mismatched conditions (especially targeting DTW metrics) and discusses approaches investigated for QUESST2014: generation of calibration side-information by a language identification system, and handling T2 and T3 queries relaxing the constraints of an exact match. All results are provided on QUESST2014 development and evaluation data.

Keywords

query processing; regression analysis; DTW metrics; QUESST2014 development; calibration side-information; channel mismatch; language identification system; query-by-example; spoken term detection technique; Calibration; Europe; acoustic keyword spotting; dynamic time warping; fusion; logistic regression; m-norm; query-by-example spoken term detection; spoken document retrieval;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech and Signal Processing (ICASSP), 2015 IEEE International Conference on

Conference_Location

South Brisbane, QLD

Type

conf

DOI

10.1109/ICASSP.2015.7179091

Filename

7179091