Publication Details

Resources and Benchmarks for Keyword Search in Spoken Audio From Low-Resource Indian Languages

NADIMPALLI Vijaya Lakshmi V., KESIRAJU Santosh, BANKA Rohith, KETHIREDDY Rashmi and GANGASHETTY Suryakanth V. Resources and Benchmarks for Keyword Search in Spoken Audio From Low-Resource Indian Languages. IEEE Access, vol. 10, no. 2022, 2022, pp. 34789-34799. ISSN 2169-3536. Available from: https://ieeexplore.ieee.org/document/9743904
Czech title
Zdroje a srovnání pro vyhledávání klíčových slov v mluveném audiu indických jazyků s malým množstvím zdrojů
Type
journal article
Language
english
Authors
Nadimpalli Vijaya Lakshmi V. (IIIT)
Kesiraju Santosh (DCGM FIT BUT)
Banka Rohith (IIIT)
Kethireddy Rashmi (IIIT)
Gangashetty Suryakanth V (IIIT)
URL
Keywords

Keyword search, low-resource languages, term-weighted value (TWV)

Abstract

This paper presents the resources and benchmarks developed for keyword search (KWS) in spoken audio from six low-resource Indian languages (from two families), namely Gujarati, Hindi, Marathi, Odia, Tamil, and Telugu. The current work on constructing keywords and building benchmark KWS systems is inspired by the popular IARPA Babel program and the subsequent works on low-resource KWS. The keywords are constructed by taking into account their properties i.e., occurrence, length, and average confusability; and their effects on the evaluation metric - the term-weighted value (TWV).We make use of freely available speech datasets, and reprocess them to create resources for KWS, thereby adding value to the existing speech resources. Four ASR-based KWS systems are built, and their performance is analyzed across the three keyword properties on all the six languages. The prepared keywords and other related resources to replicate our experiments are made available for the public.We believe that the analysis and guidelines provided in this paper will not only help the research community, but also practitioners and engineers to easily create KWS resources for newer languages, datasets, and scenarios.

Published
2022
Pages
34789-34799
Journal
IEEE Access, vol. 10, no. 2022, ISSN 2169-3536
Publisher
Institute of Electrical and Electronics Engineers
DOI
UT WoS
000778878900001
EID Scopus
BibTeX
@ARTICLE{FITPUB12952,
   author = "V. Lakshmi Vijaya Nadimpalli and Santosh Kesiraju and Rohith Banka and Rashmi Kethireddy and V Suryakanth Gangashetty",
   title = "Resources and Benchmarks for Keyword Search in Spoken Audio From Low-Resource Indian Languages",
   pages = "34789--34799",
   journal = "IEEE Access",
   volume = 10,
   number = 2022,
   year = 2022,
   ISSN = "2169-3536",
   doi = "10.1109/ACCESS.2022.3162854",
   language = "english",
   url = "https://www.fit.vut.cz/research/publication/12952"
}
Back to top