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.
Our new paper entitled "Independent Low-Rank Matrix Analysis Based on Generalized Kullback-Leibler Divergence" is now available in IEICE Trans. This is the world’s first attempt to realize convex-optimization-based ILRMA w.r.t. source modeling parameters.
https://t.co/xWAYMnY6jh