Show simple item record

dc.contributor.author
Cerutti, Gianmarco
dc.contributor.author
Andri, Renzo
dc.contributor.author
Cavigelli, Lukas
dc.contributor.author
Farella, Elisabetta
dc.contributor.author
Magno, Michele
dc.contributor.author
Benini, Luca
dc.date.accessioned
2021-02-19T09:11:30Z
dc.date.available
2021-01-03T03:32:18Z
dc.date.available
2021-02-19T09:11:30Z
dc.date.issued
2020-08
dc.identifier.isbn
978-1-4503-7053-0
en_US
dc.identifier.other
10.1145/3370748.3406588
en_US
dc.identifier.uri
http://hdl.handle.net/20.500.11850/458823
dc.description.abstract
Sound event detection (SED) is a hot topic in consumer and smart city applications. Existing approaches based on deep neural networks (DNNs) are very effective, but highly demanding in terms of memory, power, and throughput when targeting ultra-low power always-on devices. Latency, availability, cost, and privacy requirements are pushing recent IoT systems to process the data on the node, close to the sensor, with a very limited energy supply, and tight constraints on the memory size and processing capabilities precluding to run state-of-the-art DNNs. In this paper, we explore the combination of extreme quantization to a small-footprint binary neural network (BNN) with the highly energy-efficient, RISC-V-based (8+1)-core GAP8 microcontroller. Starting from an existing CNN for SED whose footprint (815 kB) exceeds the 512 kB of memory available on our platform, we retrain the network using binary filters and activations to match these memory constraints. (Fully) binary neural networks come with a natural drop in accuracy of 12-18% on the challenging ImageNet object recognition challenge compared to their equivalent full-precision baselines. This BNN reaches a 77.9% accuracy, just 7% lower than the full-precision version, with 58 kB (7.2× less) for the weights and 262 kB (2.4× less) memory in total. With our BNN implementation, we reach a peak throughput of 4.6 GMAC/s and 1.5 GMAC/s over the full network, including preprocessing with Mel bins, which corresponds to an efficiency of 67.1 GMAC/s/W and 31.3 GMAC/s/W, respectively. Compared to the performance of an ARM Cortex-M4 implementation, our system has a 10.3× faster execution time and a 51.1× higher energy-efficiency. © 2020 ACM
en_US
dc.language.iso
en
en_US
dc.publisher
ACM
en_US
dc.subject
Binary Neural Networks
en_US
dc.subject
Sound Event Detection
en_US
dc.subject
Ultra Low Power
en_US
dc.title
Sound event detection with binary neural networks on tightly power-constrained IoT devices
en_US
dc.type
Conference Paper
dc.date.published
2020-08-10
ethz.book.title
Proceedings of the ACM/IEEE International Symposium on Low Power Electronics and Design (ISLPED '20)
en_US
ethz.pages.start
19
en_US
ethz.pages.end
24
en_US
ethz.event
ACM/IEEE International Symposium on Low Power Electronics and Design (ISLPED 2020) (virtual)
en_US
ethz.event.location
Boston, MA, USA
en_US
ethz.event.date
August 10-12, 2020
en_US
ethz.notes
Due to the Coronavirus (COVID-19) the conference was conducted virtually.
en_US
ethz.identifier.scopus
ethz.publication.place
New York, NY
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.::02636 - Institut für Integrierte Systeme / Integrated Systems Laboratory::03996 - Benini, Luca / Benini, Luca
en_US
ethz.leitzahl.certified
ETH Zürich::00002 - ETH Zürich::00012 - Lehre und Forschung::00007 - Departemente::02140 - Dep. Inf.technologie und Elektrotechnik / Dep. of Inform.Technol. Electrical Eng.::02636 - Institut für Integrierte Systeme / Integrated Systems Laboratory::03996 - Benini, Luca / Benini, Luca
ethz.date.deposited
2021-01-03T03:32:22Z
ethz.source
SCOPUS
ethz.eth
yes
en_US
ethz.availability
Metadata only
en_US
ethz.rosetta.installDate
2021-02-19T09:11:41Z
ethz.rosetta.lastUpdated
2022-03-29T05:18:20Z
ethz.rosetta.versionExported
true
ethz.COinS
ctx_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.atitle=Sound%20event%20detection%20with%20binary%20neural%20networks%20on%20tightly%20power-constrained%20IoT%20devices&rft.date=2020-08&rft.spage=19&rft.epage=24&rft.au=Cerutti,%20Gianmarco&Andri,%20Renzo&Cavigelli,%20Lukas&Farella,%20Elisabetta&Magno,%20Michele&rft.isbn=978-1-4503-7053-0&rft.genre=proceeding&rft_id=info:doi/10.1145/3370748.3406588&rft.btitle=Proceedings%20of%20the%20ACM/IEEE%20International%20Symposium%20on%20Low%20Power%20Electronics%20and%20Design%20(ISLPED%20'20)
 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