Nonstationary and temporally correlated source separation using Gaussian process

Hsieh, Hsin-Lung; Chien, Jen‐Tzung

doi:10.1109/icassp.2011.5946745

Cited by 4 publications

(2 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The underdetermined separation under different number of sources and sensors could be tackled. Also, the online learning could be involved to update segment-based parameters and hyperparameters [33,34]. The evolutionary BGS-NMFs shall work for nonstationary single-channel blind source separation.…”

Section: Discussionmentioning

confidence: 99%

Bayesian group sparse learning for music source separation

Chien

Hsieh

2013

J AUDIO SPEECH MUSIC PROC.

Self Cite

View full text Add to dashboard Cite

Nonnegative matrix factorization (NMF) is developed for parts-based representation of nonnegative signals with the sparseness constraint. The signals are adequately represented by a set of basis vectors and the corresponding weight parameters. NMF has been successfully applied for blind source separation and many other signal processing systems. Typically, controlling the degree of sparseness and characterizing the uncertainty of model parameters are two critical issues for model regularization using NMF. This paper presents the Bayesian group sparse learning for NMF and applies it for single-channel music source separation. This method reconstructs the rhythmic or repetitive signal from a common subspace spanned by the shared bases for the whole signal and simultaneously decodes the harmonic or residual signal from an individual subspace consisting of separate bases for different signal segments. A Laplacian scale mixture distribution is introduced for sparse coding given a sparseness control parameter. The relevance of basis vectors for reconstructing two groups of music signals is automatically determined. A Markov chain Monte Carlo procedure is presented to infer two sets of model parameters and hyperparameters through a sampling procedure based on the conditional posterior distributions. Experiments on separating single-channel audio signals into rhythmic and harmonic source signals show that the proposed method outperforms baseline NMF, Bayesian NMF, and other group-based NMF in terms of signal-to-interference ratio.

show abstract

Section: Discussionmentioning

confidence: 99%

Bayesian group sparse learning for music source separation

Chien

Hsieh

2013

J AUDIO SPEECH MUSIC PROC.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Today, this area became a substantial branch of signal processing research [2,3]. In speech signal processing for example, there have been a variety of contributions successfully employing Bayesian techniques and probabilistic graphical models [4][5][6][7][8][9], neural networks [10,11] and deep learning [12], kernel methods [13,14], and sparsity-based techniques [15,16], to name just a few.…”

Section: Introductionmentioning

confidence: 99%

A Bayesian network viewon linear and nonlinear acoustic echo cancellation

Maas

Huemmer

Schwarz

et al. 2014

2014 IEEE China Summit &Amp; International Conference on Signal and Information Processing (ChinaSIP)

View full text Add to dashboard Cite

In this contribution, we provide a new derivation of the normalized least mean square (NLMS) algorithm from a machine learning perspective. By applying the inference rules of Bayesian networks to a linear observation model, the NLMS can be shown to arise as a modification of the Kalman filter equations. Based on a nonlinear observation model, we exemplify the benefit of the Bayesian point of view by employing the technique of particle filtering to realize a tractable algorithm for nonlinear acoustic echo cancellation. Experiments carried out on real smartphone recordings reveal the remarkable performance of the new approach.

show abstract