DocumentCode :
2357783
Title :
Stable and fast update rules for independent vector analysis based on auxiliary function technique
Author :
Ono, Nobutaka
Author_Institution :
Nat. Inst. of Inf., Tokyo, Japan
fYear :
2011
fDate :
16-19 Oct. 2011
Firstpage :
189
Lastpage :
192
Abstract :
This paper presents stable and fast update rules for independent vector analysis (IVA) based on auxiliary function technique. The algorithm consists of two alternative updates: 1) weighted covariance matrix updates and 2) demixing matrix updates, which include no tuning parameters such as step size. The monotonic decrease of the objective function at each update is guaranteed. The experimental evaluation shows that the derived update rules yield faster convergence and better results than natural gradient updates.
Keywords :
blind source separation; covariance matrices; vectors; auxiliary function technique; independent vector analysis; monotonic decrease; weighted covariance matrix updates; Conferences; Convergence; Frequency domain analysis; Source separation; Speech; Vectors; auxiliary function; blind source separation; independent vector analysis; natural gradient;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Applications of Signal Processing to Audio and Acoustics (WASPAA), 2011 IEEE Workshop on
Conference_Location :
New Paltz, NY
ISSN :
1931-1168
Print_ISBN :
978-1-4577-0692-9
Electronic_ISBN :
1931-1168
Type :
conf
DOI :
10.1109/ASPAA.2011.6082320
Filename :
6082320
Link To Document :
بازگشت