Open AccessSingle channel blind source separation based on probabilistic matrix factorisation
Author(s) -
Kim HanGyu,
Jang GilJin,
Park JeongSik,
Oh YungHwan,
Choi HoJin
Publication year - 2017
Publication title -
electronics letters
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.375
H-Index - 146
ISSN - 1350-911X
DOI - 10.1049/el.2017.2013
Subject(s) - non negative matrix factorization , source separation , blind signal separation , matrix decomposition , mathematics , algorithm , matrix (chemical analysis) , computer science , channel (broadcasting) , pattern recognition (psychology) , artificial intelligence , eigenvalues and eigenvectors , physics , computer network , materials science , quantum mechanics , composite material
A novel single channel blind source separation method based on probabilistic matrix factorisation (PMF) is proposed. Compared to the conventional non‐negative matrix factorisation (NMF) employing Euclidean distance or Kullback–Leibler divergence, PMF uses the log posterior probability as a cost function for optimising spectrum and activation matrices. Such cost function has an advantage that the hyperparameters are optimised numerically without cross‐validation. In order to apply PMF to audio source separation, both Gaussian and Laplacian priors are considered. Exponential substitution for target matrices is also proposed to guarantee the non‐negativity of the separated spectrogram. In source separation experiments, the proposed PMF‐based approach provided significantly better performance than the conventional NMF.