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

Self-supervised language models are very effective at predicting high-levelcortical responses during language comprehension. However, the best currentmodels of lower-level auditory processing in the human brain rely on eitherhand-constructed acoustic filters or representations from supervised audioneural networks. In this work, we capitalize on the progress of self-supervisedspeech representation learning (SSL) to create new state-of-the-art models ofthe human auditory system. Compared against acoustic baselines, phonemicfeatures, and supervised models, representations from the middle layers ofself-supervised models (APC, wav2vec, wav2vec 2.0, and HuBERT) consistentlyyield the best prediction performance for fMRI recordings within the auditorycortex (AC). Brain areas involved in low-level auditory processing exhibit apreference for earlier SSL model layers, whereas higher-level semantic areasprefer later layers. We show that these trends are due to the models' abilityto encode information at multiple linguistic levels (acoustic, phonetic, andlexical) along their representation depth. Overall, these results show thatself-supervised models effectively capture the hierarchy of informationrelevant to different stages of speech processing in human cortex.

Self-supervised models of audio effectively explain human cortical responses to speech