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

Curriculum learning begins to thrive in the speech enhancement area, whichdecouples the original spectrum estimation task into multiple easier sub-tasksto achieve better performance. Motivated by that, we propose a dual-branchattention-in-attention transformer dubbed DB-AIAT to handle both coarse- andfine-grained regions of the spectrum in parallel. From a complementaryperspective, a magnitude masking branch is proposed to coarsely estimate theoverall magnitude spectrum, and simultaneously a complex refining branch iselaborately designed to compensate for the missing spectral details andimplicitly derive phase information. Within each branch, we propose a novelattention-in-attention transformer-based module to replace the conventionalRNNs and temporal convolutional networks for temporal sequence modeling.Specifically, the proposed attention-in-attention transformer consists ofadaptive temporal-frequency attention transformer blocks and an adaptivehierarchical attention module, aiming to capture long-term temporal-frequencydependencies and further aggregate global hierarchical contextual information.Experimental results on Voice Bank + DEMAND demonstrate that DB-AIAT yieldsstate-of-the-art performance (e.g., 3.31 PESQ, 95.6% STOI and 10.79dB SSNR)over previous advanced systems with a relatively small model size (2.81M).

Dual-branch Attention-In-Attention Transformer for single-channel speech enhancement