Multichannel high resolution NMF for modelling convolutive mixtures of non-stationary signals in the time-frequency domain

Roland Badeau (1, 2) , Mark D. Plumbley (3)

1 LTCI - Laboratoire Traitement et Communication de l'Information
2 TSI - Département Traitement du Signal et des Images
3 Centre for Digital Music

Several probabilistic models involving latent components have been proposed for modelling time-frequency (TF) representations of audio signals such as spectrograms, notably in the nonnegative matrix factorization (NMF) literature. Among them, the recent high resolution NMF (HR-NMF) model is able to take both phases and local correlations in each frequency band into account, and its potential has been illustrated in applications such as source separation and audio inpainting. In this paper, HR-NMF is extended to multichannel signals and to convolutive mixtures. The new model can represent a variety of stationary and non-stationary signals, including autoregressive moving average (ARMA) processes and mixtures of damped sinusoids. A fast variational expectation-maximization (EM) algorithm is proposed to estimate the enhanced model. This algorithm is applied to piano signals, and proves capable of accurately modelling reverberation, restoring missing observations, and separating pure tones with close frequencies.

Mots clés: Non-stationary signal modelling; Timefrequency analysis; Nonnegative matrix factorisation; Multichannel signal analysis; Variational EM algorithm

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

Roland Badeau, Mark D. Plumbley. Multichannel high resolution NMF for modelling convolutive mixtures of non-stationary signals in the time-frequency domain. IEEE Transactions on Audio, Speech and Language Processing, 2014, 22 (11), pp.1670-1680. ⟨hal-01061578⟩