Model-based monaural sound separation by split-VQ of sinusoidal parameters | IEEE Conference Publication | IEEE Xplore