B3 Non-refereed article in a conference publication
Cheese identification based on feature extraction from rubbing sounds; 
Authors: Nurkkala, Elina; Czedik-Eysenberg, Isabella; Reuter, Christoph; Hopia, Anu; Postema, Michiel
Conference name: DAGA
Publisher: Deutsche Gesellschaft für Akustik
Publication year: 2026
Book title : Fortschritte der Akustik - DAGA 2026
Article number: 387
First page : 1130
Last page: 1132
ISBN: 978-3-939296-24-9
DOI: https://doi.org/10.71568/daga2026.387
Publication's open availability at the time of reporting: Open Access
Publication channel's open availability : Open Access publication channel
Web address : https://doi.org/10.71568/daga2026.387
When biting and chewing food, the sounds produced convey emotions that help us to identify what we are eating and to evaluate its quality. Although extensive scientific attention has been paid to fruits, vegetables, and crisps, only a sparse number of scientific manuscripts has been dedicated to cheese acoustics, despite the notorious loudness of string cheeses. The purpose of this study was to find quantitative acoustic parameters to discriminate cheese types from rubbing. Audible audio clips were collected from rubbing leipäjuusto, mozzarella, and Gouda cheese samples with a wetted blade in a controlled low-noise environment. Twelve hundred clips were processed automatically to extract psychoacoustic features which included harmonic loudness, nontempered energy ratio, and dynamic complexity. It was found that, based on psychoacoustic features, cheese types can be discriminated with near-perfect accuracy. In addition, from a psychoacoustic point of view, leipäjuusto can be classified as “annoying yet interesting”.