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

Residual learning is a recently proposed learning framework to facilitate thetraining of very deep neural networks. Residual blocks or units are made of aset of stacked layers, where the inputs are added back to their outputs withthe aim of creating identity mappings. In practice, such identity mappings areaccomplished by means of the so-called skip or residual connections. However,multiple implementation alternatives arise with respect to where such skipconnections are applied within the set of stacked layers that make up aresidual block. While ResNet architectures for image classification usingconvolutional neural networks (CNNs) have been widely discussed in theliterature, few works have adopted ResNet architectures so far for 1D audioclassification tasks. Thus, the suitability of different residual block designsfor raw audio classification is partly unknown. The purpose of this paper is toanalyze and discuss the performance of several residual block implementationswithin a state-of-the-art CNN-based architecture for end-to-end audioclassification using raw audio waveforms. For comparison purposes, we analyzeas well the performance of the residual blocks under a similar 2D architectureusing a conventional time-frequency audio represen-tation as input. The resultsshow that the achieved accuracy is considerably dependent, not only on thespecific residual block implementation, but also on the selected inputnormalization.

On the performance of residual block design alternatives in convolutional neural networks for end-to-end audio classification