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

Modern smart glasses leverage advanced audio sensing and machine learningtechnologies to offer real-time transcribing and captioning services,considerably enriching human experiences in daily communications. However, suchsystems frequently encounter challenges related to environmental noises,resulting in degradation to speech recognition and speaker change detection. Toimprove voice quality, this work investigates directional source separationusing the multi-microphone array. We first explore multiple beamformers toassist source separation modeling by strengthening the directional propertiesof speech signals. In addition to relying on predetermined beamformers, weinvestigate neural beamforming in multi-channel source separation,demonstrating that automatic learning directional characteristics effectivelyimproves separation quality. We further compare the ASR performance leveragingseparated outputs to noisy inputs. Our results show that directional sourceseparation benefits ASR for the wearer but not for the conversation partner.Lastly, we perform the joint training of the directional source separation andASR model, achieving the best overall ASR performance.

Directional Source Separation for Robust Speech Recognition on Smart Glasses