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/

Global: Zendy Plus annual

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

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Sound Event Detection and Localization (SELD) constitutes a complex task thatdepends on extensive multichannel audio recordings with annotated sound eventsand their respective locations. In this paper, we introduce a self-supervisedapproach for SELD adapted from the pre-training methodology of wav2vec 2.0,which learns representations directly from raw audio data, eliminating the needfor supervision. By applying this approach to SELD, we can leverage asubstantial amount of unlabeled 3D audio data to learn robust representationsof sound events and their locations. Our method comprises two primary stages:pre-training and fine-tuning. In the pre-training phase, unlabeled 3D audiodatasets are utilized to train our w2v-SELD model, capturing intricatehigh-level features and contextual information inherent in audio signals.Subsequently, in the fine-tuning stage, a smaller dataset with labeled SELDdata fine-tunes the pre-trained model. Experimental results on benchmarkdatasets demonstrate the effectiveness of the proposed self-supervised approachfor SELD. The model surpasses baseline systems provided with the datasets andachieves competitive performance comparable to state-of-the-art supervisedmethods. The code and pre-trained parameters of our w2v-SELD model areavailable in this repository.

w2v-SELD: A Sound Event Localization and Detection Framework for Self-Supervised Spatial Audio Pre-Training