Show simple item record

dc.contributor.author
Chen, Qinyu
dc.contributor.author
Chang, Yaoxing
dc.contributor.author
Kim, Kwantae
dc.contributor.author
Gao, Chang
dc.contributor.author
Liu, Shih-Chii
dc.date.accessioned
2023-09-12T06:23:20Z
dc.date.available
2023-09-09T05:58:10Z
dc.date.available
2023-09-12T06:23:20Z
dc.date.issued
2023
dc.identifier.isbn
978-1-6654-5109-3
en_US
dc.identifier.other
10.1109/ISCAS46773.2023.10181602
en_US
dc.identifier.uri
http://hdl.handle.net/20.500.11850/630732
dc.description.abstract
Keyword spotting (KWS) is an important task on edge low-power audio devices. A typical edge KWS system consists of a front-end feature extractor which outputs mel-scale frequency cepstral coefficients (MFCC) features followed by a back-end neural network classifier. KWS edge designs aim for the best power-performance-area metrics. This work proposes an area-efficient ultra-low-power time-domain infinite impulse response (IIR) filter-based feature extractor for a KWS system. It uses a serial architecture, and the architecture is further optimized for a low-cost computing structure and mixed-precision bit selection of the IIR coefficients while maintaining good KWS accuracy. Using a 65nm process technology and a back-end neural network classifier, this simulated feature extractor has an area of 0.02mm2 and achieves 3.3 mu W @ 1.2V, and achieves 92.5% accuracy on a 10-keyword, 12-class KWS task using the GSCD dataset.
en_US
dc.language.iso
en
en_US
dc.publisher
IEEE
en_US
dc.subject
Keyword spotting (KWS)
en_US
dc.subject
infinite impulse response (IIR)
en_US
dc.subject
hardware acceleration
en_US
dc.subject
long short-term memory
en_US
dc.title
An Area-Efficient Ultra-Low-Power Time-Domain Feature Extractor for Edge Keyword Spotting
en_US
dc.type
Conference Paper
dc.date.published
2023-07-21
ethz.book.title
IEEE ISCAS 2023 Symposium Proceedings
en_US
ethz.size
5 p.
en_US
ethz.event
IEEE International Symposium on Circuits and Systems (ISCAS 2023)
en_US
ethz.event.location
Monterey, CA, USA
en_US
ethz.event.date
May 21-25, 2023
en_US
ethz.identifier.wos
ethz.publication.place
Piscataway, NJ
en_US
ethz.publication.status
published
en_US
ethz.leitzahl
ETH Zürich::00002 - ETH Zürich::00012 - Lehre und Forschung::00007 - Departemente::02140 - Dep. Inf.technologie und Elektrotechnik / Dep. of Inform.Technol. Electrical Eng.::02533 - Institut für Neuroinformatik / Institute of Neuroinformatics
ethz.date.deposited
2023-09-09T05:58:16Z
ethz.source
WOS
ethz.eth
yes
en_US
ethz.availability
Metadata only
en_US
ethz.rosetta.installDate
2023-09-12T06:23:22Z
ethz.rosetta.lastUpdated
2024-02-03T03:26:17Z
ethz.rosetta.versionExported
true
ethz.COinS
ctx_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.atitle=An%20Area-Efficient%20Ultra-Low-Power%20Time-Domain%20Feature%20Extractor%20for%20Edge%20Keyword%20Spotting&rft.date=2023&rft.au=Chen,%20Qinyu&Chang,%20Yaoxing&Kim,%20Kwantae&Gao,%20Chang&Liu,%20Shih-Chii&rft.isbn=978-1-6654-5109-3&rft.genre=proceeding&rft_id=info:doi/10.1109/ISCAS46773.2023.10181602&rft.btitle=IEEE%20ISCAS%202023%20Symposium%20Proceedings
 Search print copy at ETH Library

Files in this item

FilesSizeFormatOpen in viewer

There are no files associated with this item.

Publication type

Show simple item record