Audio-Based Onset Detection applied to Chewing Cycle Segmentation

Publication Type Conference Paper
Authors David Kopyto, Rui Zhang, Oliver Amft
Title Audio-Based Onset Detection applied to Chewing Cycle Segmentation
Abstract In this paper we compare three onset detection algorithms for acoustic chewing cycle detection, which is a basic step in eating detection and automated dietary monitoring. We introduce a spectral flux algorithm that uses the spectrogram of a chewing sequence to compute a novelty function. Furthermore, beat tracking, in particular the notion of a predominant local pulse is introduced. We compare the two algorithms to a baseline energy-based segmentation in a chewing dataset with seven participants consuming pieces of six different foods, including in total 9818 annotated chewing cycles. Best performance was achieved for the beat tracking algorithm with 83% F-measure after leave-one-participant-out cross validation.
Date September 21, 2021
Proceedings Title ACM International Symposium on Wearable Computers
Conference Name ISWC '21
Place Virtual Conference
Publisher Association for Computing Machinery
Pages 124–128
Series ISWC '21
DOI 10.1145/3460421.3478819
ISBN 978-1-4503-8462-9
URL Publisher's website
Accessed 2021-09-21
Library Catalog ACM Digital Library
Full Text PDF
Friedrich-Alexander-Universität Erlangen-Nürnberg