Hybrid Multichannel Signal Separation Using Supervised Nonnegative Matrix Factorization with Spectrogram Restoration

In this paper, we propose a new hybrid method that concatenates directional clustering and advanced nonnegative matrix factorization (NMF) for the purpose of the specific sound extraction from the multichannel music signal. Multichannel music signal separation technology is aimed to extract a specific target signal from observed multichannel signals that contain multiple instrumental sounds. In the previous studies, various methods using NMF have been proposed, but they remain many problems, e.g., poor convergence in update rules in NMF and lack of robustness. To solve these problems, we propose a new supervised NMF (SNMF) with spectrogram restoration and its hybrid method that concatenates the proposed SNMF after directional clustering. Via extrapolation of supervised spectral bases, the proposed SNMF attempts both target signal separation and reconstruction of the lost target components, which are generated by preceding directional clustering. In addition, we theoretically reveal the trade-off between separation and extrapolation abilities and propose a new scheme for multi-divergence, where optimal divergence can be automatically changed in each time frame according to the local spatial conditions. The results of an evaluation experiment show that our proposed hybrid method outperforms the conventional music signal separation methods.