Improving Sound Event Detection with Auxiliary Foreground-Background Classification and Domain Adaptation - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2021

Improving Sound Event Detection with Auxiliary Foreground-Background Classification and Domain Adaptation

Abstract

In this paper we provide two methods that improve the detection of sound events in domestic environments. First, motivated by the broad categorization of domestic sounds as foreground or background events according to their spectro-temporal structure, we propose to learn a foreground-background classifier jointly with the sound event classifier in a multi-task fashion to improve the generalization of the latter. Second, while the semi-supervised learning capability adopted for training sound event detection systems with synthetic labeled data and unlabeled or partially labeled real data aims to learn invariant representations for both domains, there is still a gap in performance when testing such systems on real environments. To further reduce this data mismatch, we propose a domain adaptation strategy that aligns the empirical distributions of the feature representations of active and inactive frames of synthetic and real recordings via optimal transport. We show that these two approaches lead to enhanced detection performance in terms of the event-based macro F1-score on the DESED dataset.
Fichier principal
Vignette du fichier
DCASE_Workshop_2021.pdf (583.78 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-03387778 , version 1 (20-10-2021)

Identifiers

  • HAL Id : hal-03387778 , version 1

Cite

Michel Olvera, Emmanuel Vincent, Gilles Gasso. Improving Sound Event Detection with Auxiliary Foreground-Background Classification and Domain Adaptation. DCASE 2021 - 6th Workshop on Detection and Classification of Acoustic Scenes and Events, Nov 2021, Virtual, Spain. ⟨hal-03387778⟩
185 View
309 Download

Share

Gmail Mastodon Facebook X LinkedIn More