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

Modern neural speech enhancement models usually include various forms ofphase information in their training loss terms, either explicitly orimplicitly. However, these loss terms are typically designed to reduce thedistortion of phase spectrum values at specific frequencies, which ensures theydo not significantly affect the quality of the enhanced speech. In this paper,we propose an effective phase reconstruction strategy for neural speechenhancement that can operate in noisy environments. Specifically, we introducea phase continuity loss that considers relative phase variations across thetime and frequency axes. By including this phase continuity loss in astate-of-the-art neural speech enhancement system trained with reconstructionloss and a number of magnitude spectral losses, we show that our proposedmethod further improves the quality of enhanced speech signals over thebaseline, especially when training is done jointly with a magnitude spectrumloss.

Phase Continuity: Learning Derivatives of Phase Spectrum for Speech Enhancement