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

Textless self-supervised speech models have grown in capabilities in recentyears, but the nature of the linguistic information they encode has not yetbeen thoroughly examined. We evaluate the extent to which these models' learnedrepresentations align with basic representational distinctions made by humans,focusing on a set of phonetic (low-level) and phonemic (more abstract)contrasts instantiated in word-initial stops. We find that robustrepresentations of both phonetic and phonemic distinctions emerge in earlylayers of these models' architectures, and are preserved in the principalcomponents of deeper layer representations. Our analyses suggest two sourcesfor this success: some can only be explained by the optimization of the modelson speech data, while some can be attributed to these models' high-dimensionalarchitectures. Our findings show that speech-trained HuBERT derives a low-noiseand low-dimensional subspace corresponding to abstract phonologicaldistinctions.

Probing self-supervised speech models for phonetic and phonemic information: a case study in aspiration