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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Self-supervised speech representations are known to encode both speaker andphonetic information, but how they are distributed in the high-dimensionalspace remains largely unexplored. We hypothesize that they are encoded inorthogonal subspaces, a property that lends itself to simple disentanglement.Applying principal component analysis to representations of two predictivecoding models, we identify two subspaces that capture speaker and phoneticvariances, and confirm that they are nearly orthogonal. Based on this property,we propose a new speaker normalization method which collapses the subspace thatencodes speaker information, without requiring transcriptions. Probingexperiments show that our method effectively eliminates speaker information andoutperforms a previous baseline in phone discrimination tasks. Moreover, theapproach generalizes and can be used to remove information of unseen speakers.

Self-supervised Predictive Coding Models Encode Speaker and Phonetic Information in Orthogonal Subspaces