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

Unsupervised anomalous sound detection (ASD) aims to identify anomaloussounds by learning the features of normal operational sounds and sensing theirdeviations. Recent approaches have focused on the self-supervised taskutilizing the classification of normal data, and advanced models have shownthat securing representation space for anomalous data is important throughrepresentation learning yielding compact intra-class and well-separatedintra-class distributions. However, we show that conventional approaches oftenfail to ensure sufficient intra-class compactness and exhibit angular disparitybetween samples and their corresponding centers. In this paper, we propose atraining technique aimed at ensuring intra-class compactness and increasing theangle gap between normal and abnormal samples. Furthermore, we present anarchitecture that extracts features for important temporal regions, enablingthe model to learn which time frames should be emphasized or suppressed.Experimental results demonstrate that the proposed method achieves the bestperformance giving 0.90%, 0.83%, and 2.16% improvement in terms of AUC, pAUC,and mAUC, respectively, compared to the state-of-the-art method on DCASE 2020Challenge Task2 dataset.

Noisy-ArcMix: Additive Noisy Angular Margin Loss Combined With Mixup Anomalous Sound Detection