Urban sound classifcation using neural networks on embedded FPGAs
View/ Open
Impact
Scholar |
Other documents of the author: BELLOCH, JOSE A.; Coronado, Raul; Valls, Oscar; Amor, Rocío del; Leon, German; Naranjo, Valery; Dolz, Manuel F.; Amor-Martin, Adrian; Piñero, Gema
Metadata
Show full item recordcomunitat-uji-handle:10234/9
comunitat-uji-handle2:10234/7036
comunitat-uji-handle3:10234/8620
comunitat-uji-handle4:
INVESTIGACIONMetadata
Title
Urban sound classifcation using neural networks on embedded FPGAsAuthor (s)
Date
2024-03-01Publisher
SpringerISSN
0920-8542; 1573-0484Bibliographic citation
Belloch, J.A., Coronado, R., Valls, O. et al. Urban sound classification using neural networks on embedded FPGAs. J Supercomput (2024). https://doi.org/10.1007/s11227-024-05947-8Type
info:eu-repo/semantics/articleVersion
info:eu-repo/semantics/publishedVersionSubject
Abstract
Sound classifcation using neural networks has recently produced very accurate
results. A large number of diferent applications use this type of sound classifers
such as controlling and monitoring the type of activity ... [+]
Sound classifcation using neural networks has recently produced very accurate
results. A large number of diferent applications use this type of sound classifers
such as controlling and monitoring the type of activity in a city or identifying different types of animals in natural environments. While traditional acoustic processing applications have been developed on high-performance computing platforms
equipped with expensive multi-channel audio interfaces, the Internet of Things (IoT)
paradigm requires the use of more fexible and energy-efcient systems. Although
software-based platforms exist for implementing general-purpose neural networks,
they are not optimized for sound classifcation, wasting energy and computational
resources. In this work, we have used FPGAs to develop an ad hoc system where
only the hardware needed for our application is synthesized, resulting in faster and
more energy-efcient circuits. The results show that our developments are accelerated by a factor of 35 compared to a software-based implementation on a Raspberry
Pi. [-]
Related data
No additional data or materials available.Funder Name
CRUE-CSIC agreement with Springer Nature | NextGenerationEU/PRTR | Ministerio de Ciencia, Innovación y Universidades | Generalitat Valenciana
Project code
MCIN/AEI/10.13039/501100011033 | PID2020-113656RB | PID2021-124280OB-C21 | PID2022-137048OA-C43 | TED2021-131401B-C21 (DIPSY-AI) | TED2021-131401A-C22 (DIPSYTECH) | CIDEXG/2022/013
Project title or grant
ERDF A way of making Europe
Rights
© The Author(s) 2024
info:eu-repo/semantics/openAccess
info:eu-repo/semantics/openAccess
This item appears in the folowing collection(s)
- ICC_Articles [427]