Stereo sound event localization and detection based on PSELDnet pretraining and BiMamba sequence modeling

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Gao, Wenmiao, Xiao, Yang
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915346668060672
author Gao, Wenmiao
Xiao, Yang
author_facet Gao, Wenmiao
Xiao, Yang
contents Pre-training methods have achieved significant performance improvements in sound event localization and detection (SELD) tasks, but existing Transformer-based models suffer from high computational complexity. In this work, we propose a stereo sound event localization and detection system based on pre-trained PSELDnet and bidirectional Mamba sequence modeling. We replace the Conformer module with a BiMamba module and introduce asymmetric convolutions to more effectively model the spatiotemporal relationships between time and frequency dimensions. Experimental results demonstrate that the proposed method achieves significantly better performance than the baseline and the original PSELDnet with Conformer decoder architecture on the DCASE2025 Task 3 development dataset, while also reducing computational complexity. These findings highlight the effectiveness of the BiMamba architecture in addressing the challenges of the SELD task.
format Preprint
id arxiv_https___arxiv_org_abs_2506_13455
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Stereo sound event localization and detection based on PSELDnet pretraining and BiMamba sequence modeling
Gao, Wenmiao
Xiao, Yang
Audio and Speech Processing
Sound
Pre-training methods have achieved significant performance improvements in sound event localization and detection (SELD) tasks, but existing Transformer-based models suffer from high computational complexity. In this work, we propose a stereo sound event localization and detection system based on pre-trained PSELDnet and bidirectional Mamba sequence modeling. We replace the Conformer module with a BiMamba module and introduce asymmetric convolutions to more effectively model the spatiotemporal relationships between time and frequency dimensions. Experimental results demonstrate that the proposed method achieves significantly better performance than the baseline and the original PSELDnet with Conformer decoder architecture on the DCASE2025 Task 3 development dataset, while also reducing computational complexity. These findings highlight the effectiveness of the BiMamba architecture in addressing the challenges of the SELD task.
title Stereo sound event localization and detection based on PSELDnet pretraining and BiMamba sequence modeling
topic Audio and Speech Processing
Sound
url https://arxiv.org/abs/2506.13455