A Sequential Self Teaching Approach for Improving Generalization in Sound Event Recognition

TL;DR

This paper introduces a sequential self-teaching method for sound event recognition that enhances model generalization, especially with noisy or weakly labeled data, demonstrated by up to 9% performance gains on Audioset.

Contribution

It proposes a novel stage-wise learning approach that improves generalization and transferability in sound event recognition models under challenging data conditions.

Findings

Up to 9% performance improvement on Audioset.

Enhanced transferability of learned knowledge.

Better generalization in adverse data scenarios.

Abstract

An important problem in machine auditory perception is to recognize and detect sound events. In this paper, we propose a sequential self-teaching approach to learning sounds. Our main proposition is that it is harder to learn sounds in adverse situations such as from weakly labeled and/or noisy labeled data, and in these situations a single stage of learning is not sufficient. Our proposal is a sequential stage-wise learning process that improves generalization capabilities of a given modeling system. We justify this method via technical results and on Audioset, the largest sound events dataset, our sequential learning approach can lead to up to 9% improvement in performance. A comprehensive evaluation also shows that the method leads to improved transferability of knowledge from previously trained models, thereby leading to improved generalization capabilities on transfer learning tasks.