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

Humans possess the remarkable ability to selectively attend to a singlespeaker amidst competing voices and background noise, known as selectiveauditory attention. Recent studies in auditory neuroscience indicate a strongcorrelation between the attended speech signal and the corresponding brain'selicited neuronal activities, which the latter can be measured using affordableand non-intrusive electroencephalography (EEG) devices. In this study, wepresent NeuroHeed, a speaker extraction model that leverages EEG signals toestablish a neuronal attractor which is temporally associated with the speechstimulus, facilitating the extraction of the attended speech signal in acocktail party scenario. We propose both an offline and an online NeuroHeed,with the latter designed for real-time inference. In the online NeuroHeed, weadditionally propose an autoregressive speaker encoder, which accumulates pastextracted speech signals for self-enrollment of the attended speakerinformation into an auditory attractor, that retains the attentional momentumover time. Online NeuroHeed extracts the current window of the speech signalswith guidance from both attractors. Experimental results demonstrate thatNeuroHeed effectively extracts brain-attended speech signals, achieving highsignal quality, excellent perceptual quality, and intelligibility in atwo-speaker scenario.

NeuroHeed: Neuro-Steered Speaker Extraction using EEG Signals