Class-Incremental Learning for Sound Event Localization and Detection

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Pandey, Ruchi, Mulimani, Manjunath, Politis, Archontis, Mesaros, Annamaria
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866915026244206592
author Pandey, Ruchi
Mulimani, Manjunath
Politis, Archontis
Mesaros, Annamaria
author_facet Pandey, Ruchi
Mulimani, Manjunath
Politis, Archontis
Mesaros, Annamaria
contents This paper investigates the feasibility of class-incremental learning (CIL) for Sound Event Localization and Detection (SELD) tasks. The method features an incremental learner that can learn new sound classes independently while preserving knowledge of old classes. The continual learning is achieved through a mean square error-based distillation loss to minimize output discrepancies between subsequent learners. The experiments are conducted on the TAU-NIGENS Spatial Sound Events 2021 dataset, which includes 12 different sound classes and demonstrate the efficacy of proposed method. We begin by learning 8 classes and introduce the 4 new classes at next stage. After the incremental phase, the system is evaluated on the full set of learned classes. Results show that, for this realistic dataset, our proposed method successfully maintains baseline performance across all metrics.
format Preprint
id arxiv_https___arxiv_org_abs_2411_12830
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Class-Incremental Learning for Sound Event Localization and Detection
Pandey, Ruchi
Mulimani, Manjunath
Politis, Archontis
Mesaros, Annamaria
Audio and Speech Processing
This paper investigates the feasibility of class-incremental learning (CIL) for Sound Event Localization and Detection (SELD) tasks. The method features an incremental learner that can learn new sound classes independently while preserving knowledge of old classes. The continual learning is achieved through a mean square error-based distillation loss to minimize output discrepancies between subsequent learners. The experiments are conducted on the TAU-NIGENS Spatial Sound Events 2021 dataset, which includes 12 different sound classes and demonstrate the efficacy of proposed method. We begin by learning 8 classes and introduce the 4 new classes at next stage. After the incremental phase, the system is evaluated on the full set of learned classes. Results show that, for this realistic dataset, our proposed method successfully maintains baseline performance across all metrics.
title Class-Incremental Learning for Sound Event Localization and Detection
topic Audio and Speech Processing
url https://arxiv.org/abs/2411.12830