UCIL: An Unsupervised Class Incremental Learning Approach for Sound Event Detection

2024-07-04 · Yang Xiao, Rohan Kumar Das

This work explores class-incremental learning (CIL) for sound event detection (SED), advancing adaptability towards real-world scenarios. CIL's success in domains like computer vision inspired our SED-tailored method, addressing the unique challenges of diverse and complex audio environments. Our approach employs an independent unsupervised learning framework with a distillation loss function to integrate new sound classes while preserving the SED model consistency across incremental tasks. We further enhance this framework with a sample selection strategy for unlabeled data and a balanced exemplar update mechanism, ensuring varied and illustrative sound representations. Evaluating various continual learning methods on the DCASE 2023 Task 4 dataset, we find that our research offers insights into each method's applicability for real-world SED systems that can have newly added sound classes. The findings also delineate future directions of CIL in dynamic audio settings.

📄 PDF Abstract BibTeX arXiv:2407.03657

Code (0)

등록된 구현이 없습니다.

Tasks

class-incremental learning Class Incremental Learning Continual LearningEvent DetectionIncremental Learning Sound Event Detectionunsupervised class-incremental learning

Exploiting Fine-Grained Prototype Distribution for Boosting Unsupervised Class Incremental Learning

2024-08-19 · Jiaming Liu, Hongyuan Liu, Zhili Qin, Wei Han 외

The dynamic nature of open-world scenarios has attracted more attention to class incremental learning (CIL). However, existing CIL methods typically presume the availability of complete ground-truth labels throughout the…

class-incremental learning Class Incremental Learning Incremental Learning unsupervised class-incremental learning

Class-Incremental Learning for Sound Event Localization and Detection

2024-11-19 · Ruchi Pandey, Manjunath Mulimani, Archontis Politis, Annamaria Mesaros

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 independent…

class-incremental learning Class Incremental Learning Continual Learning Incremental Learning+1

Unsupervised Incremental Learning and Prediction of Music Signals

2015-02-02 · Ricard Marxer, Hendrik Purwins

A system is presented that segments, clusters and predicts musical audio in an unsupervised manner, adjusting the number of (timbre) clusters instantaneously to the audio input. A sequence learning algorithm adapts its s…

Clustering Incremental Learning Onset Detection Prediction

Incremental Learning of Acoustic Scenes and Sound Events

2023-02-28 · Manjunath Mulimani, Annamaria Mesaros

In this paper, we propose a method for incremental learning of two distinct tasks over time: acoustic scene classification (ASC) and audio tagging (AT). We use a simple convolutional neural network (CNN) model as an incr…

Acoustic Scene Classification Audio Tagging Incremental Learning Knowledge Distillation+1

Walking the Web of Concept-Class Relationships in Incrementally Trained Interpretable Models

2025-02-27 · Susmit Agrawal, Deepika Vemuri, Sri Siddarth Chakaravarthy P, Vineeth N. Balasubramanian

Concept-based methods have emerged as a promising direction to develop interpretable neural networks in standard supervised settings. However, most works that study them in incremental settings assume either a static con…