Skip to main navigation Skip to search Skip to main content

Sound Event Detection on A Single MCU: Lightweight, Low-power and Portable

  • School of Electronics and Information Engineering, Harbin Institute of Technology
  • Marine Design & Research Institute of China

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Sound event detection (SED) in Internet of things (IoT) domain always face the challenges of strict power consumption constraints. To achieve ultra-low power, many SED systems use complex analog circuit to extract sound feature and construct algorithm, which discourages the portability and generality of the algorithm to be deployed on other IoT devices, and also takes up more spaces. In this paper, we propose a light-weight SED algorithm based on long-term speech diversion (LTSD) feature extraction and support vector machine(SVM) classifier to realize SED fully on a single Microcontroller Unit (MCU) without depending on any extra analog frontend. The algorithm achieves accuracy rates of 86.3%, better than other lightweight SED algorithms. Onboard deployment is realized on a single MCU CC1350, achieving 88.2% recognition accuracy within 3.125 μs speed, which only occupies 33.4KB flash and 9.94kB SRAM, and consumes 23.1mW power, reflecting its ability to be portable on other tightly memory-constrained and low-power IoT devices.

Original languageEnglish
Title of host publication2022 International Conference on Sensing, Measurement and Data Analytics in the Era of Artificial Intelligence, ICSMD 2022 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781665492812
DOIs
StatePublished - 2022
Externally publishedYes
Event3rd International Conference on Sensing, Measurement and Data Analytics in the Era of Artificial Intelligence, ICSMD 2022 - Harbin, China
Duration: 22 Dec 202224 Dec 2022

Publication series

Name2022 International Conference on Sensing, Measurement and Data Analytics in the Era of Artificial Intelligence, ICSMD 2022 - Proceedings

Conference

Conference3rd International Conference on Sensing, Measurement and Data Analytics in the Era of Artificial Intelligence, ICSMD 2022
Country/TerritoryChina
CityHarbin
Period22/12/2224/12/22

Keywords

  • Internet of Things
  • Long-Term Speech Diversion
  • Sound Event Detection
  • Support Vector Machine

Fingerprint

Dive into the research topics of 'Sound Event Detection on A Single MCU: Lightweight, Low-power and Portable'. Together they form a unique fingerprint.

Cite this