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

Recent advances in neural network based acoustic modelling have shownsignificant improvements in automatic speech recognition (ASR) performance. Inorder for acoustic models to be able to handle large acoustic variability,large amounts of labeled data is necessary, which are often expensive toobtain. This paper explores the application of adversarial training to learnfeatures from raw speech that are invariant to acoustic variability. Thisacoustic variability is referred to as a domain shift in this paper. Theexperimental study presented in this paper leverages the architecture of DomainAdversarial Neural Networks (DANNs) [1] which uses data from two differentdomains. The DANN is a Y-shaped network that consists of a multi-layer CNNfeature extractor module that is common to a label (senone) classifier and aso-called domain classifier. The utility of DANNs is evaluated on multipledatasets with domain shifts caused due to differences in gender and speakeraccents. Promising empirical results indicate the strength of adversarialtraining for unsupervised domain adaptation in ASR, thereby emphasizing theability of DANNs to learn domain invariant features from raw speech.

Adversarial Learning of Raw Speech Features for Domain Invariant Speech Recognition