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

Articulatory representation learning is the fundamental research in modelingneural speech production system. Our previous work has established a deepparadigm to decompose the articulatory kinematics data into gestures, whichexplicitly model the phonological and linguistic structure encoded with humanspeech production mechanism, and corresponding gestural scores. We continuewith this line of work by raising two concerns: (1) The articulators areentangled together in the original algorithm such that some of the articulatorsdo not leverage effective moving patterns, which limits the interpretability ofboth gestures and gestural scores; (2) The EMA data is sparsely sampled fromarticulators, which limits the intelligibility of learned representations. Inthis work, we propose a novel articulatory representation decompositionalgorithm that takes the advantage of guided factor analysis to derive thearticulatory-specific factors and factor scores. A neural convolutive matrixfactorization algorithm is then employed on the factor scores to derive the newgestures and gestural scores. We experiment with the rtMRI corpus that capturesthe fine-grained vocal tract contours. Both subjective and objective evaluationresults suggest that the newly proposed system delivers the articulatoryrepresentations that are intelligible, generalizable, efficient andinterpretable.

Articulatory Representation Learning Via Joint Factor Analysis and Neural Matrix Factorization