Minimum-volume regularized ILRMA for blind audio source separation

Multichannel blind audio source separation aims to recover the latent sources from their multichannel mixture without priors. A state-of-the-art blind audio source separation method called independent low-rank matrix analysis (ILRMA) unified independent vector analysis (IVA) and nonnegative matrix factorization (NMF). However the spectra matrix produced from NMF may not find a compact representation. Also the matrix may not guarantee that each source is identifiable. To address the problem here we propose a modified blind audio source separation method that enhances the identifiability of the source model. It combines ILRMA with a new penalty term named minimum volume regularization The proposed method is optimized by standard majorization-minimization framework based multiplication updating rule which ensures the stability of convergence. Experimental results demonstrate the effectiveness of the proposed method compared with AuxIVA, MNMF and ILRMA.