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

Prediction of late reverberation component using multi-channel linearprediction (MCLP) in short-time Fourier transform (STFT) domain is an effectivemeans to enhance reverberant speech. Traditionally, a speech power spectraldensity (PSD) weighted prediction error (WPE) minimization approach is used toestimate the prediction filters. The method is sensitive to the estimate of thedesired signal PSD. In this paper, we propose a deep neural network (DNN) basednon-linear estimate for the desired signal PSD. An auto encoder trained onclean speech STFT coefficients is used as the desired signal prior. We exploretwo different architectures based on (i) fully-connected (FC) feed-forward, and(ii) recurrent long short-term memory (LSTM) layers. Experiments using realroom impulse responses show that the LSTM-DNN based PSD estimate performsbetter than the traditional methods for late reverb suppression.

LSTM based AE-DNN constraint for better late reverb suppression in multi-channel LP formulation