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

Gaussian process (GP) audio source separation is a time-domain approach thatcircumvents the inherent phase approximation issue of spectrogram basedmethods. Furthermore, through its kernel, GPs elegantly incorporate priorknowledge about the sources into the separation model. Despite these compellingadvantages, the computational complexity of GP inference scales cubically withthe number of audio samples. As a result, source separation GP models have beenrestricted to the analysis of short audio frames. We introduce an efficientapplication of GPs to time-domain audio source separation, without compromisingperformance. For this purpose, we used GP regression, together with spectralmixture kernels, and variational sparse GPs. We compared our method withLD-PSDTF (positive semi-definite tensor factorization), KL-NMF(Kullback-Leibler non-negative matrix factorization), and IS-NMF (Itakura-SaitoNMF). Results show that the proposed method outperforms these techniques.

Sparse Gaussian Process Audio Source Separation Using Spectrum Priors in the Time-Domain