Contrastive Loss Based Frame-wise Feature disentanglement for Polyphonic Sound Event Detection

Autor:	Guan, Yadong, Han, Jiqing, Song, Hongwei, Song, Wenjie, Zheng, Guibin, Zheng, Tieran, He, Yongjun
Rok vydání:	2024
Předmět:	Computer Science - Sound Electrical Engineering and Systems Science - Audio and Speech Processing
Druh dokumentu:	Working Paper
Popis:	Overlapping sound events are ubiquitous in real-world environments, but existing end-to-end sound event detection (SED) methods still struggle to detect them effectively. A critical reason is that these methods represent overlapping events using shared and entangled frame-wise features, which degrades the feature discrimination. To solve the problem, we propose a disentangled feature learning framework to learn a category-specific representation. Specifically, we employ different projectors to learn the frame-wise features for each category. To ensure that these feature does not contain information of other categories, we maximize the common information between frame-wise features within the same category and propose a frame-wise contrastive loss. In addition, considering that the labeled data used by the proposed method is limited, we propose a semi-supervised frame-wise contrastive loss that can leverage large amounts of unlabeled data to achieve feature disentanglement. The experimental results demonstrate the effectiveness of our method. Comment: accepted by icassp2024
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2401.05850 Zobrazit plný text záznamu View this record from Arxiv