Blind Source Separation Exploiting Higher-Order Frequency Dependencies

Kim, Taesu; Attias, Hagai; Lee, Sang Yup; Lee, Te-Won

doi:10.1109/tasl.2006.872618

Cited by 384 publications

(382 citation statements)

References 20 publications

Supporting

Mentioning

380

Contrasting

Unclassified

Order By: Relevance

“…in which we assume the density function (13). By setting α = 0 and b = 1/2, the nonlinear function becomes simpler as…”

Section: Complex-valued Icamentioning

confidence: 99%

“…Among them, we consider the frequency-domain approach [6][7][8][9][10][11][12][13][14] where we apply a short-time Fourier transform (STFT) to the sensor observations xj(t). If we use a sufficiently long frame for STFT to cover the main part of the impulse responses h jk , the convolutive mixture (1) can be approximated well with an instantaneous mixture at each frequency f :…”

Section: Introductionmentioning

confidence: 99%

“…The ambiguities should be aligned properly so that the separated signals that originate from the same source are grouped together. This problem is known as the permutation problem of frequencydomain BSS, and various methods [6][7][8][9][10][11][12][13][14] have been proposed for its solution. Section 4 discusses a strategy that exploits the mutual dependence of bin-wise separated signals across frequencies [8][9][10]14].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

MLSP 2007 Data Analysis Competition: Frequency-Domain Blind Source Separation for Convolutive Mixtures of Speech/Audio Signals

Sawada

Araki

Makino

2007

2007 IEEE Workshop on Machine Learning for Signal Processing

View full text Add to dashboard Cite

This paper describes the frequency-domain approach to the blind source separation of speech/audio signals that are convolutively mixed in a real room environment. With the application of shorttime Fourier transforms, convolutive mixtures in the time domain can be approximated as multiple instantaneous mixtures in the frequency domain. We employ complex-valued independent component analysis (ICA) to separate the mixtures in each frequency bin. Then, the permutation ambiguity of the ICA solutions should be aligned so that the separated signals are constructed properly in the time domain. We propose a permutation alignment method based on clustering the activity sequences of the frequency binwise separated signals. We achieved the overall winner status of MLSP 2007 Data Analysis Competition based on the presented method.

show abstract

“…in which we assume the density function (13). By setting α = 0 and b = 1/2, the nonlinear function becomes simpler as…”

Section: Complex-valued Icamentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

MLSP 2007 Data Analysis Competition: Frequency-Domain Blind Source Separation for Convolutive Mixtures of Speech/Audio Signals

Sawada

Araki

Makino

2007

2007 IEEE Workshop on Machine Learning for Signal Processing

View full text Add to dashboard Cite

show abstract

“…The rapid expansion of urbanization is increasingly putting impacts on speech blind separation, and the traditional methods of speech blind separation have exposed their limitations under new circumstances, with the application of fast fixed-point independent vector analysis into it, the unban speech blind separation is instantly developing unprecedentedly [1]. Independent component analysis is a famous and has been widely used in the "blind source separation" of the algorithm.…”

Section: Introductionmentioning

confidence: 99%

Research of City Engineering Speech Blind Separation Algorithms Based on Independent Vector Analysis

Zhao¹,

Li²

2012

Proceedings of the 2nd International Conference on Electronic and Mechanical Engineering and Information Technology (2012)

View full text Add to dashboard Cite

Abstract. Aiming at the rapid expansion of urbanization, This paper proposed a new independent vector analysis model, it is an extension of ICA for independent multivariate speech where the sources are mixed component-wise. Formulation that can solve the frequency-domain blind source separation problem, which is very useful in directly applying Newton's method to a contrast function of complex-valued variables. Fast fixed-point independent vector analysis blind source separation algorithms are derived .The real speech signal are used to demonstrate the good converge and accuracy of the algorithms by simulating.

show abstract

“…In [10] a method is proposed for convolutive blind source separation in reverberative environments using a frequency domain approach with sources having variance (scale) dependency across frequencies. Typically in the frequency domain approach to blind deconvolution, the "permutation problem" arises when the signals are unmixed separately at each frequency.…”

Section: Introductionmentioning

confidence: 99%

Modeling and Estimation of Dependent Subspaces with Non-radially Symmetric and Skewed Densities

Palmer

Kreutz-Delgado

Rao

et al.

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. We extend the Gaussian scale mixture model of dependent subspace source densities to include non-radially symmetric densities using Generalized Gaussian random variables linked by a common variance. We also introduce the modeling of skew in source densities and subspaces using a generalization of the Normal Variance-Mean mixture model. We give closed form expressions for subspace likelihoods and parameter updates in the EM algorithm.

show abstract

Blind Source Separation Exploiting Higher-Order Frequency Dependencies

Cited by 384 publications

References 20 publications

MLSP 2007 Data Analysis Competition: Frequency-Domain Blind Source Separation for Convolutive Mixtures of Speech/Audio Signals

MLSP 2007 Data Analysis Competition: Frequency-Domain Blind Source Separation for Convolutive Mixtures of Speech/Audio Signals

Research of City Engineering Speech Blind Separation Algorithms Based on Independent Vector Analysis

Modeling and Estimation of Dependent Subspaces with Non-radially Symmetric and Skewed Densities

Contact Info

Product

Resources

About