A Multi-grained based Attention Network for Semi-supervised Sound Event Detection

TL;DR

This paper introduces MGA-Net, a semi-supervised sound event detection model that uses multi-grained attention and hybrid convolution to improve detection accuracy, especially with limited data.

Contribution

The paper proposes MGA-Net with a residual hybrid convolution block and a multi-grained attention module, enhancing feature extraction and temporal resolution for better sound event detection.

Findings

Outperforms state-of-the-art methods on benchmark datasets.

Achieves 53.27% and 56.96% macro F1 scores on validation and test sets.

Demonstrates effectiveness of the spatial shift data augmentation.

Abstract

Sound event detection (SED) is an interesting but challenging task due to the scarcity of data and diverse sound events in real life. This paper presents a multi-grained based attention network (MGA-Net) for semi-supervised sound event detection. To obtain the feature representations related to sound events, a residual hybrid convolution (RH-Conv) block is designed to boost the vanilla convolution's ability to extract the time-frequency features. Moreover, a multi-grained attention (MGA) module is designed to learn temporal resolution features from coarse-level to fine-level. With the MGA module,the network could capture the characteristics of target events with short- or long-duration, resulting in more accurately determining the onset and offset of sound events. Furthermore, to effectively boost the performance of the Mean Teacher (MT) method, a spatial shift (SS) module as a data perturbation mechanism is introduced to increase the diversity of data. Experimental results show that the MGA-Net outperforms the published state-of-the-art competitors, achieving 53.27% and 56.96% event-based macro F1 (EB-F1) score, 0.709 and 0.739 polyphonic sound detection score (PSDS) on the validation and public set respectively.