Phase recovery in NMF for audio source separation: an insightful benchmark

Paul Magron (1) , Roland Badeau (1) , Bertrand David (1)

1 TSI - Département Traitement du Signal et des Images

Nonnegative Matrix Factorization (NMF) is a powerful tool for decomposing mixtures of audio signals in the Time-Frequency (TF) domain. In applications such as source separation, the phase recovery for each extracted component is a major issue since it often leads to audible artifacts. In this paper, we present a methodology for evaluating various NMF-based source separation techniques involving phase reconstruction. For each model considered, a comparison between two approaches (blind separation without prior information and oracle separation with supervised model learning) is performed, in order to inquire about the room for improvement for the estimation methods. Experimental results show that the High Resolution NMF (HRNMF) model is particularly promising, because it is able to take phases and correlations over time into account with a great expressive power.

Mots clés: Nonnegative matrix factorization; audio source separation; time-frequency analysis; phase reconstruction

Domaines: Traitement du signal et de l'image [eess.SP]

Paul Magron, Roland Badeau, Bertrand David. Phase recovery in NMF for audio source separation: an insightful benchmark. 40th International Conference on Acoustics, Speech and Signal Processing (ICASSP), Apr 2015, Brisbane, Australia. pp.81-85. ⟨hal-01110032⟩