Acoustic scene classification from few examples

Ivan Bocharov, Tjalling Tjalkens, Bert de Vries

Onderzoeksoutput: Hoofdstuk in Boek/Rapport/CongresprocedureConferentiebijdrageAcademicpeer review

1 Citaat (Scopus)
411 Downloads (Pure)

Samenvatting

In order to personalize the behavior of hearing aid devices in different acoustic environments, we need to develop personalized acoustic scene classifiers. Since we cannot afford to burden an individual hearing aid user with the task to collect a large acoustic database, we aim instead to train a scene classifier on just one (or maximally a few) in-situ recorded acoustic waveform of a few seconds duration per scene. In this paper we develop such a”one-shot” personalized scene classifier, based on a Hidden Semi-Markov model. The presented classifier consistently outperforms a more classical Dynamic-Time-Warping-Nearest-Neighbor classifier, and correctly classifies acoustic scenes about twice as well as a (random) chance classifier after training on just one recording of 10 seconds duration per scene.

Originele taal-2Engels
Titel2018 26th European Signal Processing Conference, EUSIPCO 2018
Plaats van productiePiscataway
UitgeverijInstitute of Electrical and Electronics Engineers
Pagina's862-866
Aantal pagina's5
Volume2018-September
ISBN van elektronische versie9789082797015
ISBN van geprinte versie978-90-827970-1-5
DOI's
StatusGepubliceerd - sep. 2018
Evenement26th European Signal Processing Conference, EUSIPCO 2018 - Rome, Italië
Duur: 3 sep. 20187 sep. 2018

Congres

Congres26th European Signal Processing Conference, EUSIPCO 2018
Verkorte titelEUSIPCO 2018
Land/RegioItalië
StadRome
Periode3/09/187/09/18

Vingerafdruk

Duik in de onderzoeksthema's van 'Acoustic scene classification from few examples'. Samen vormen ze een unieke vingerafdruk.

Citeer dit