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

Deep Learning based Automatic Speech Recognition (ASR) models are verysuccessful, but hard to interpret. To gain better understanding of howArtificial Neural Networks (ANNs) accomplish their tasks, introspection methodshave been proposed. Adapting such techniques from computer vision to speechrecognition is not straight-forward, because speech data is more complex andless interpretable than image data. In this work, we introduceGradient-adjusted Neuron Activation Profiles (GradNAPs) as means to interpretfeatures and representations in Deep Neural Networks. GradNAPs arecharacteristic responses of ANNs to particular groups of inputs, whichincorporate the relevance of neurons for prediction. We show how to utilizeGradNAPs to gain insight about how data is processed in ANNs. This includesdifferent ways of visualizing features and clustering of GradNAPs to compareembeddings of different groups of inputs in any layer of a given network. Wedemonstrate our proposed techniques using a fully-convolutional ASR model.

Gradient-Adjusted Neuron Activation Profiles for Comprehensive Introspection of Convolutional Speech Recognition Models