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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Acoustic scene classification (ASC) has been approached in the last yearsusing deep learning techniques such as convolutional neural networks orrecurrent neural networks. Many state-of-the-art solutions are based on imageclassification frameworks and, as such, a 2D representation of the audio signalis considered for training these networks. Finding the most suitable audiorepresentation is still a research area of interest. In this paper, differentlog-Mel representations and combinations are analyzed. Experiments show thatthe best results are obtained using the harmonic and percussive components plusthe difference between left and right stereo channels, (L-R). On the otherhand, it is a common strategy to ensemble different models in order to increasethe final accuracy. Even though averaging different model predictions is acommon choice, an exhaustive analysis of different ensemble techniques has notbeen presented in ASC problems. In this paper, geometric and arithmetic meanplus the Ordered Weighted Averaging (OWA) operator are studied as aggregationoperators for the output of the different models of the ensemble. Finally, thework carried out in this paper is highly oriented towards real-timeimplementations. In this context, as the number of applications for audioclassification on edge devices is increasing exponentially, we also analyzedifferent network depths and efficient solutions for aggregating ensemblepredictions.

CNN depth analysis with different channel inputs for Acoustic Scene Classification