Open AccessMusic/Voice Separation Based on Kernel Back‐Fitting Using Weighted β ‐Order MMSE Estimation
Author(s) -
Kim HyoungGook,
Kim Jin Young
Publication year - 2016
Publication title -
etri journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.295
H-Index - 46
eISSN - 2233-7326
pISSN - 1225-6463
DOI - 10.4218/etrij.16.0115.0256
Subject(s) - kernel (algebra) , estimator , spectrogram , separation (statistics) , computer science , algorithm , source separation , minimum mean square error , mean squared error , mathematics , speech recognition , artificial intelligence , statistics , machine learning , combinatorics
Recent developments in the field of separation of mixed signals into music/voice components have attracted the attention of many researchers. Recently, iterative kernel back‐fitting, also known as kernel additive modeling, was proposed to achieve good results for music/voice separation. To obtain minimum mean square error (MMSE) estimates of short‐time Fourier transforms of sources, generalized spatial Wiener filtering (GW) is typically used. In this paper, we propose an advanced music/voice separation method that utilizes a generalized weighted β ‐order MMSE estimation (WbE) based on iterative kernel back‐fitting (KBF). In the proposed method, WbE is used for the step of mixed music signal separation, while KBF permits kernel spectrogram model fitting at each iteration. Experimental results show that the proposed method achieves better separation performance than GW and existing Bayesian estimators.