Detailed Information

Cited 0 time in webofscience Cited 0 time in scopus
Metadata Downloads

Discriminatory and Orthogonal Feature Learning for Noise Robust Keyword Spotting

Authors
Kim, DonghyeonKo, KyungdeukHan, David K.Ko, Hanseok
Issue Date
2022
Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
Keywords
Measurement; Computational modeling; Feature extraction; Mathematical models; Convolution; Training; Euclidean distance; Keyword Spotting; robustness; metric learning
Citation
IEEE SIGNAL PROCESSING LETTERS, v.29, pp.1913 - 1917
Indexed
SCIE
SCOPUS
Journal Title
IEEE SIGNAL PROCESSING LETTERS
Volume
29
Start Page
1913
End Page
1917
URI
https://scholar.korea.ac.kr/handle/2021.sw.korea/145567
DOI
10.1109/LSP.2022.3203911
ISSN
1070-9908
Abstract
Keyword Spotting (KWS) is an essential component in a smart device for alerting the system when a user prompts it with a command. As these devices are typically constrained by computational and energy resources, the KWS model should be designed with a small footprint. In our previous work, we developed lightweight dynamic filters which extract a robust feature map within a noisy environment. The learning variables of the dynamic filter are jointly optimized with KWS weights by using Cross-Entropy (CE) loss. CE loss alone, however, is not sufficient for high performance when the SNR is low. In order to train the network for more robust performance in noisy environments, we introduce the LOw Variant Orthogonal (LOVO) loss. The LOVO loss is composed of a triplet loss applied on the output of the dynamic filter, a spectral norm-based orthogonal loss, and an inner class distance loss applied in the KWS model. These losses are particularly useful in encouraging the network to extract discriminatory features in unseen noise environments.
Files in This Item
There are no files associated with this item.
Appears in
Collections
College of Engineering > School of Electrical Engineering > 1. Journal Articles

qrcode

Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.

Related Researcher

Researcher Ko, Han seok photo

Ko, Han seok
공과대학 (전기전자공학부)
Read more

Altmetrics

Total Views & Downloads

BROWSE