Publication Details

End-to-End Open Vocabulary Keyword Search

YUSUF Bolaji, GOK Alican, GUNDOGDU Batuhan and SARAÇLAR Murat. End-to-End Open Vocabulary Keyword Search. In: Proceedings Interspeech 2021. Brno: International Speech Communication Association, 2021, pp. 4388-4392. ISSN 1990-9772. Available from: https://www.isca-speech.org/archive/interspeech_2021/yusuf21_interspeech.html
Czech title
End-to-End detekce klíčových slov s neomezeným slovníkem
Type
conference paper
Language
english
Authors
Yusuf Bolaji (DCGM FIT BUT)
Gok Alican (UBOGAZ)
Gundogdu Batuhan (UBOGAZ)
Saraçlar Murat (UBOGAZ)
URL
Keywords

keyword search, spoken term detection

Abstract

Recently, neural approaches to spoken content retrieval have become popular. However, they tend to be restricted in their vocabulary or in their ability to deal with imbalanced test settings. These restrictions limit their applicability in keyword search, where the set of queries is not known beforehand, and where the system should return not just whether an utterance contains a query but the exact location of any such occurrences. In this work, we propose a model directly optimized for keyword search. The model takes a query and an utterance as input and returns a sequence of probabilities for each frame of the utterance of the query having occurred in that frame. Experiments show that the proposed model not only outperforms similar end-to-end models on a task where the ratio of positive and negative trials is artificially balanced, but it is also able to deal with the far more challenging task of keyword search with its inherent imbalance. Furthermore, using our system to rescore the outputs an LVCSR-based keyword search system leads to significant improvements on the latter.

Published
2021
Pages
4388-4392
Journal
Proceedings of Interspeech - on-line, vol. 2021, no. 8, ISSN 1990-9772
Proceedings
Proceedings Interspeech 2021
Conference
Interspeech Conference, Brno, CZ
Publisher
International Speech Communication Association
Place
Brno, CZ
DOI
UT WoS
000841879504096
EID Scopus
BibTeX
@INPROCEEDINGS{FITPUB12612,
   author = "Bolaji Yusuf and Alican Gok and Batuhan Gundogdu and Murat Sara\c{c}lar",
   title = "End-to-End Open Vocabulary Keyword Search",
   pages = "4388--4392",
   booktitle = "Proceedings Interspeech 2021",
   journal = "Proceedings of Interspeech - on-line",
   volume = 2021,
   number = 8,
   year = 2021,
   location = "Brno, CZ",
   publisher = "International Speech Communication Association",
   ISSN = "1990-9772",
   doi = "10.21437/Interspeech.2021-1399",
   language = "english",
   url = "https://www.fit.vut.cz/research/publication/12612"
}
Back to top