Independent Low-Rank Matrix Analysis Based on Generalized Kullback-Leibler Divergence

Regular Section

Shinichi MOGAMI, Yoshiki MITSUI, Norihiro TAKAMUNE, Daichi KITAMURA, Hiroshi SARUWATARI, Yu TAKAHASHI, Kazunobu KONDO, Hiroaki NAKAJIMA, Hirokazu KAMEOKA

Author information

Keywords: blind source separation, nonnegative matrix factorization, Poisson distribution, Kullback-Leibler divergence

JOURNAL FREE ACCESS

2019 Volume E102.A Issue 2 Pages 458-463

DOI https://doi.org/10.1587/transfun.E102.A.458

Details

Abstract

In this letter, we propose a new blind source separation method, independent low-rank matrix analysis based on generalized Kullback-Leibler divergence. This method assumes a time-frequency-varying complex Poisson distribution as the source generative model, which yields convex optimization in the spectrogram estimation. The experimental evaluation confirms the proposed method's efficacy.

Corresponding author

Register with J-STAGE for free!