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

Phase information has a significant impact on speech perceptual quality andintelligibility. However, existing speech enhancement methods encounterlimitations in explicit phase estimation due to the non-structural nature andwrapping characteristics of the phase, leading to a bottleneck in enhancedspeech quality. To overcome the above issue, in this paper, we proposedMP-SENet, a novel Speech Enhancement Network which explicitly enhancesMagnitude and Phase spectra in parallel. The proposed MP-SENet adopts a codecarchitecture in which the encoder and decoder are bridged by time-frequencyTransformers along both time and frequency dimensions. The encoder aims toencode time-frequency representations derived from the input distortedmagnitude and phase spectra. The decoder comprises dual-stream magnitude andphase decoders, directly enhancing magnitude and wrapped phase spectra byincorporating a magnitude estimation architecture and a phase parallelestimation architecture, respectively. To train the MP-SENet model effectively,we define multi-level loss functions, including mean square error andperceptual metric loss of magnitude spectra, anti-wrapping loss of phasespectra, as well as mean square error and consistency loss of short-timecomplex spectra. Experimental results demonstrate that our proposed MP-SENetexcels in high-quality speech enhancement across multiple tasks, includingspeech denoising, dereverberation, and bandwidth extension. Compared toexisting phase-aware speech enhancement methods, it successfully avoids thebidirectional compensation effect between the magnitude and phase, leading to abetter harmonic restoration. Notably, for the speech denoising task, theMP-SENet yields a state-of-the-art performance with a PESQ of 3.60 on thepublic VoiceBank+DEMAND dataset.

Explicit Estimation of Magnitude and Phase Spectra in Parallel for High-Quality Speech Enhancement