You are currently logged in as an
Institutional Subscriber.
If you would like to logout,
please click on the button below.
Home / Publications / E-library page
Only AES members and Institutional Journal Subscribers can download
General purpose audio blind source separation algorithms have to deal with a large dynamic range for the different sources to be separated. In our algorithm the mixture is separated into single notes. These notes are clustered to construct the melodies played by the active sources. The non-negative matrix factorization (NMF) leads to good results in clustering the notes according to spectral features. The cost function for the NMF is controlled by a parameter beta. The choice of beta depends on the dynamic difference of the sources. The novelty of this paper is to propose a simple classifier to adjust the parameter beta to current dynamic ranges for increasing the separation quality.
Author (s): Spiertz, Martin; Gnann, Volke;
Affiliation:
Institut für Nachrichtentechnik, RWTH Aachen University, Aachen, Germany
(See document for exact affiliation information.)
AES Convention: 128
Paper Number:8130
Publication Date:
2010-05-06
Import into BibTeX
Session subject:
Audio Processing—Music and Speech Signal Processing
Permalink: https://aes2.org/publications/elibrary-page/?id=15427
(238KB)
Click to purchase paper as a non-member or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member Join the AES. If you need to check your member status, login to the Member Portal.
Spiertz, Martin; Gnann, Volke;; 2010; Beta Divergence for Clustering in Monaural Blind Source Separation [PDF]; Institut für Nachrichtentechnik, RWTH Aachen University, Aachen, Germany; Paper 8130; Available from: https://aes2.org/publications/elibrary-page/?id=15427
Spiertz, Martin; Gnann, Volke;; Beta Divergence for Clustering in Monaural Blind Source Separation [PDF]; Institut für Nachrichtentechnik, RWTH Aachen University, Aachen, Germany; Paper 8130; 2010 Available: https://aes2.org/publications/elibrary-page/?id=15427