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

Automatic Pronunciation Assessment (APA) plays a vital role inComputer-assisted Pronunciation Training (CAPT) when evaluating a secondlanguage (L2) learner's speaking proficiency. However, an apparent downside ofmost de facto methods is that they parallelize the modeling process throughoutdifferent speech granularities without accounting for the hierarchical andlocal contextual relationships among them. In light of this, a novelhierarchical approach is proposed in this paper for multi-aspect andmulti-granular APA. Specifically, we first introduce the notion of sup-phonemesto explore more subtle semantic traits of L2 speakers. Second, a depth-wiseseparable convolution layer is exploited to better encapsulate the localcontext cues at the sub-word level. Finally, we use a score-restraint attentionpooling mechanism to predict the sentence-level scores and optimize thecomponent models with a multitask learning (MTL) framework. Extensiveexperiments carried out on a publicly-available benchmark dataset, viz.speechocean762, demonstrate the efficacy of our approach in relation to somecutting-edge baselines.

A Hierarchical Context-aware Modeling Approach for Multi-aspect and Multi-granular Pronunciation Assessment