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

Independent deeply learned matrix analysis (IDLMA) is one of thestate-of-the-art multichannel audio source separation methods using the sourcepower estimation based on deep neural networks (DNNs). The DNN-based powerestimation works well for sounds having timbres similar to the DNN trainingdata. However, the sounds to which IDLMA is applied do not always have suchtimbres, and the timbral mismatch causes the performance degradation of IDLMA.To tackle this problem, we focus on a blind source separation counterpart ofIDLMA, independent low-rank matrix analysis. It uses nonnegative matrixfactorization (NMF) as the source model, which can capture source spectralcomponents that only appear in the target mixture, using the low-rank structureof the source spectrogram as a clue. We thus extend the DNN-based source modelto encompass the NMF-based source model on the basis of the product-of-expertconcept, which we call the product of source models (PoSM). For the proposedPoSM-based IDLMA, we derive a computationally efficient parameter estimationalgorithm based on an optimization principle called themajorization-minimization algorithm. Experimental evaluations show theeffectiveness of the proposed method.

Multichannel Audio Source Separation with Independent Deeply Learned Matrix Analysis Using Product of Source Models