Negatif olmayan matris ayriştirma ile tek-kanaldan algisal ses ayriştirma

Translated title of the contribution: Perceptual single-channel audio source separation by non-negative matrix factorization

Serap Kirbiz*, Bilge Günsel

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

3 Citations (Scopus)

Abstract

This paper proposes a single-channel audio source decomposition method that integrates perceptual quality criteria into source separation. Unlike the existing methods, the proposed method applies a perceptually weighted non-negative matrix factorization on log-frequency spectrogram of the mixed signal. The weights are adaptively calculated for each critical band based on a perceptual model described by ITU-R BS. 1387 perceptual quality standard. It is shown that the proposed adaptive weighting scheme significantly improves the quality of audio sources estimated by minimizing the weighted divergence between the observed log-frequency spectrogram and the model.

Translated title of the contributionPerceptual single-channel audio source separation by non-negative matrix factorization
Original languageTurkish
Title of host publication2009 IEEE 17th Signal Processing and Communications Applications Conference, SIU 2009
Pages416-419
Number of pages4
DOIs
Publication statusPublished - 2009
Event2009 IEEE 17th Signal Processing and Communications Applications Conference, SIU 2009 - Antalya, Turkey
Duration: 9 Apr 200911 Apr 2009

Publication series

Name2009 IEEE 17th Signal Processing and Communications Applications Conference, SIU 2009

Conference

Conference2009 IEEE 17th Signal Processing and Communications Applications Conference, SIU 2009
Country/TerritoryTurkey
CityAntalya
Period9/04/0911/04/09

Fingerprint

Dive into the research topics of 'Perceptual single-channel audio source separation by non-negative matrix factorization'. Together they form a unique fingerprint.

Cite this