Iterative knowledge distillation in r-cnns for weakly-labeled semi-supervised sound event detection

Autoren Khaled Koutini
Hamid Eghbal-zadeh
Gerhard Widmer
Editoren M. D. Plumbley
C. Kroos
J. P. Bello
G. Richard
D.P.W. Ellis
A. Mesaros
Titel Iterative knowledge distillation in r-cnns for weakly-labeled semi-supervised sound event detection
Buchtitel Proceedings of the Detection and Classification of Acoustic Scenes and Events 2018 Workshop (DCASE2018)
Typ in Konferenzband
Verlag Tampere University
ISBN 978-952-15-4262-6
Monat November
Jahr 2018
Seiten 173-177
SCCH ID# 18096
Abstract

In this paper, we present our approach used for the CP-JKU submission in Task 4 of the DCASE-2018 Challenge. We propose a novel iterative knowledge distillation technique for weakly-labeled semi-supervised event detection using neural networks, specifically Recurrent Convolutional Neural Networks (R-CNNs). R-CNNs are used to tag the unlabeled data and predict strong labels. Further, we use the R-CNN strong pseudo-labels on the training datasets and train new models after applying label-smoothing techniques on the strong pseudo-labels. Our proposed approach significantly improved the performance of the baseline, achieving the event-based f-measure of 40.86% compared to 15.11% event-based f-measure of the baseline in the provided test set from the development dataset.