English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

Sound event localization and detection (SELD) involves sound event detection(SED) and direction of arrival (DoA) estimation tasks. SED mainly relies ontemporal dependencies to distinguish different sound classes, while DoAestimation depends on spatial correlations to estimate source directions. Thispaper addresses the need to simultaneously extract spatial-temporal informationin audio signals to improve SELD performance. A novel block, the sliding-windowgraph-former (SwG-former), is designed to learn temporal context information ofsound events based on their spatial correlations. The SwG-former blocktransforms audio signals into a graph representation and constructs graphvertices to capture higher abstraction levels for spatial correlations. It usesdifferent-sized sliding windows to adapt various sound event durations andaggregates temporal features with similar spatial information whileincorporating multi-head self-attention (MHSA) to model global information.Furthermore, as the cornerstone of message passing, a robust Conv2dAgg functionis proposed and embedded into the block to aggregate the features of neighborvertices. As a result, a SwG-former model, which stacks the SwG-former blocks,demonstrates superior performance compared to recent advanced SELD models. TheSwG-former block is also integrated into the event-independent network version2 (EINV2), called SwG-EINV2, which surpasses the state-of-the-art (SOTA)methods under the same acoustic environment.

SwG-former: A Sliding-Window Graph Convolutional Network for  Simultaneous Spatial-Temporal Information Extraction in Sound Event  Localization and Detection