Couple Learning for semi-supervised sound event detection

TL;DR

This paper introduces a Couple Learning approach that enhances semi-supervised sound event detection by combining a well-trained model with a Mean Teacher model, improving pseudo-label quality and overall performance.

Contribution

It proposes a novel Couple Learning method that integrates a well-trained model with Mean Teacher, boosting semi-supervised sound event detection accuracy.

Findings

Achieved 44.25% F1-score on DCASE2020 Task 4

Outperformed baseline with 32.39% F1-score

Validated effectiveness through Variable Order Input experiment

Abstract

The recently proposed Mean Teacher method, which exploits large-scale unlabeled data in a self-ensembling manner, has achieved state-of-the-art results in several semi-supervised learning benchmarks. Spurred by current achievements, this paper proposes an effective Couple Learning method that combines a well-trained model and a Mean Teacher model. The suggested pseudo-labels generated model (PLG) increases strongly- and weakly-labeled data to improve the Mean Teacher method-s performance. Moreover, the Mean Teacher-s consistency cost reduces the noise impact in the pseudo-labels introduced by detection errors. The experimental results on Task 4 of the DCASE2020 challenge demonstrate the superiority of the proposed method, achieving about 44.25% F1-score on the public evaluation set, significantly outperforming the baseline system-s 32.39%. At the same time, we also propose a simple and effective experiment called the Variable Order Input (VOI) experiment, which proves the significance of the Couple Learning method. Our developed Couple Learning code is available on GitHub.