Complex NMF: A new sparse representation for acoustic signals

Author

Kameoka, Hirokazu ; Ono, Nobutaka ; Kashino, Kunio ; Sagayama, Shigeki

Author_Institution

NTT Commun. Sci. Labs., NTT Corp., Atsugi

fYear

2009

fDate

19-24 April 2009

Firstpage

3437

Lastpage

3440

Abstract

This paper presents a new sparse representation for acoustic signals which is based on a mixing model defined in the complex-spectrum domain (where additivity holds), and allows us to extract recurrent patterns of magnitude spectra that underlie observed complex spectra and the phase estimates of constituent signals. An efficient iterative algorithm is derived, which reduces to the multiplicative update algorithm for non-negative matrix factorization developed by Lee under a particular condition.

Keywords

acoustic signal processing; audio coding; iterative methods; matrix decomposition; signal representation; spectral analysis; acoustic signal; complex spectra; complex-spectrum domain; constituent signal phase estimation; iterative algorithm; magnitude spectra; multiplicative update algorithm; nonnegative matrix factorization; recurrent pattern extraction; sparse coding; sparse signal representation; Independent component analysis; Information science; Iterative algorithms; Laboratories; Phase estimation; Principal component analysis; Sparse matrices; Speech analysis; Speech processing; Vectors; Sparse signal representation; data-driven approach; non-negative matrix factorization; sparse coding;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech and Signal Processing, 2009. ICASSP 2009. IEEE International Conference on

Conference_Location

Taipei

ISSN

1520-6149

Print_ISBN

978-1-4244-2353-8

Electronic_ISBN

1520-6149

Type

conf

DOI

10.1109/ICASSP.2009.4960364

Filename

4960364