IEICE Trans - Bayesian Nonparametric Approach to Blind Separation of Infinitely Many Sparse Sources

Bayesian Nonparametric Approach to Blind Separation of Infinitely Many Sparse Sources

Hirokazu KAMEOKA  Misa SATO  Takuma ONO  Nobutaka ONO  Shigeki SAGAYAMA

Publication
IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences   Vol.E96-A    No.10    pp.1928-1937
Publication Date: 2013/10/01
Online ISSN: 1745-1337
DOI: 10.1587/transfun.E96.A.1928
Print ISSN: 0916-8508
Type of Manuscript: Special Section PAPER (Special Section on Sparsity-aware Signal Processing)
Category:
Keyword:
underdetermined blind signal separation,  speech,  sparseness,  permutation alignment,  Bayesian nonparametrics,  direction of arrival,  Dirichlet process,  stick-breaking construction,  variational inference,

Full Text: PDF(3.8MB)>>

Summary:
This paper deals with the problem of underdetermined blind source separation (BSS) where the number of sources is unknown. We propose a BSS approach that simultaneously estimates the number of sources, separates the sources based on the sparseness of speech, estimates the direction of arrival of each source, and performs permutation alignment. We confirmed experimentally that reasonably good separation was obtained with the present method without specifying the number of sources.