MTDA-HSED: Mutual-Assistance Tuning and Dual-Branch Aggregating for Heterogeneous Sound Event Detection
Preprint
Abstract
Sound Event Detection (SED) plays a vital role in comprehending and perceiving acoustic scenes. Previous methods have demonstrated impressive capabilities. However, they are deficient in learning features of complex scenes from heterogeneous dataset. In this paper, we introduce a novel dual-branch architecture named Mutual-Assistance Tuning and Dual-Branch Aggregating for Heterogeneous Sound Event Detection (MTDA-HSED). The MTDA-HSED architecture employs the Mutual-Assistance Audio Adapter (M3A) to effectively tackle the multi-scenario problem and uses the Dual-Branch Mid-Fusion (DBMF) module to tackle the multi-granularity problem. Specifically, M3A is integrated into the BEATs block as an adapter to improve the BEATs' performance by fine-tuning it on the multi-scenario dataset. The DBMF module connects BEATs and CNN branches, which facilitates the deep fusion of information from the BEATs and the CNN branches. Experimental results show that the proposed methods exceed the baseline of mpAUC by \textbf{$5\%$} on the DESED and MAESTRO Real datasets. Code is available at https://github.com/Visitor-W/MTDA.Keywords
All Related Versions
This publication has 11 references indexed in Scilit:
- AST-SED: An Effective Sound Event Detection Method Based on Audio Spectrogram Transformer2023
- Strong Labeling of Sound Events Using Crowdsourced Weak Labels and Annotator Competence EstimationIEEE/ACM Transactions on Audio, Speech, and Language Processing, 2023
- Enriched CNN-Transformer Feature Aggregation Networks for Super-Resolution2023
- Gun identification from gunshot audios for secure public places using transformer learningScientific Reports, 2022
- Threshold Independent Evaluation of Sound Event Detection Scores2022
- Sound Event Detection: A tutorialIEEE Signal Processing Magazine, 2021
- Towards Duration Robust Weakly Supervised Sound Event DetectionIEEE/ACM Transactions on Audio, Speech, and Language Processing, 2021
- Visual Object Detector for Cow Sound Event DetectionIEEE Access, 2020
- A Framework for the Robust Evaluation of Sound Event Detection2020
- SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition2019