A generalization of blind source separation algorithms for convolutive mixtures based on second-order statistics

Buchner, Herbert; Aichner, Robert; Kellermann, Walter

doi:10.1109/tsa.2004.838775

Cited by 231 publications

(185 citation statements)

References 27 publications

Supporting

Mentioning

176

Contrasting

Unclassified

Order By: Relevance

“…Firstly, we see that the disjointness maximization algorithm (10) is better than the amplitude-modulation decorrelation algorithm given in (15). Secondly, the results in Table I show that the new approach (10) performs better than the methods in [7] and [8], and slightly better than the one in [13].…”

Section: A Simulation Results For the New Algorithmmentioning

confidence: 88%

“…In accordance with the frequency-domain integrated objective function in [13] and similar second-order statistics-based approaches found in [14] and [15], we define the following frequency-domain integrated objective function that will be optimized according to the time-domain parameters of the separation system: (9) where . The operator diag[.]…”

Section: Bss Based On the Maximization Of Disjointness Of Subbanmentioning

confidence: 99%

“…The learning rules corresponding to (15) and its implementation version (16) are (15) (16) where is the same as that defined in (10), but is a diagonal matrix whose diagonal entries are the th column of the matrix . Both algorithms (10) and (15) are essentially decorrelation approaches.…”

Section: Wherementioning

confidence: 99%

See 2 more Smart Citations

Convolutive Blind Source Separation Based on Disjointness Maximization of Subband Signals

Mei

Mertins

2008

IEEE Signal Process. Lett.

View full text Add to dashboard Cite

Abstract-The concept of disjoint component analysis (DCA) is based on the fact that different speech or audio signals are typically more disjoint than mixtures of them. This letter studies the problem of blind separation of convolutive mixtures through the subband-wise maximization of the disjointness of time-frequency representations of the signals. In our approach, we first define a frequency-dependent measure representing the closeness to disjointness of a group of subband signals. Then, this frequency-dependent measure is integrated to form an objective function that only depends on the time-domain parameters of the separation system. Lastly, an efficient natural-gradient-based learning rule is developed for the update of the separation-system coefficients.Index Terms-Convolutive blind source separation, disjointness maximization.

show abstract

Section: A Simulation Results For the New Algorithmmentioning

confidence: 88%

Section: Bss Based On the Maximization Of Disjointness Of Subbanmentioning

confidence: 99%

See 1 more Smart Citation

Convolutive Blind Source Separation Based on Disjointness Maximization of Subband Signals

Mei

Mertins

2008

IEEE Signal Process. Lett.

View full text Add to dashboard Cite

show abstract

“…Localization and separation results are compared to the state-of-the-art EM-based sound source separation and localization algorithm MESSL [14] in table 1. The version MESSL-G used includes a garbage component and ILD priors to better account for reverberations and is reported to outperform four methods in reverberant conditions in terms of separation [3,16,4,17]. Note that this algorithm, as well as the vast majority of existing source localization methods [3,4,5,7,9,10], do not make use a training set 2D source locations and hence they only provide time difference of arrival for each source, i.e., frontal azimuth and no elevation.…”

Section: Methodsmentioning

confidence: 99%

Variational EM for binaural sound-source separation and localization

Deleforge

Forbes

Horaud

2013

2013 IEEE International Conference on Acoustics, Speech and Signal Processing

View full text Add to dashboard Cite

The sound-source separation and localization (SSL) problems are addressed within a unified formulation. Firstly, a mapping between white-noise source locations and binaural cues is estimated. Secondly, SSL is solved via Bayesian inversion of this mapping in the presence of multiple sparse-spectrum emitters (such as speech), noise and reverberations. We propose a variational EM algorithm which is described in detail together with initialization and convergence issues. Extensive real-data experiments show that the method outperforms the state-of-the-art both in separation and localization (azimuth and elevation).

show abstract

“…Convolutive ICA models have traditionally been in use to perform blind separation of acoustically recorded signals into individual sources (e.g., several speakers, or speaker and noise source, [17,2,8]). The convolutive mixing model comes into play through room acoustics, where the signal of each speaker has to be convolved with the room's impulse response from the speaker to each of the microphones to obtain the signal at each microphone.…”

Section: Methodsmentioning

confidence: 99%

Spatio-temporal dynamics in fMRI recordings revealed with complex independent component analysis

et al. 2006

View full text Add to dashboard Cite

Independent component analysis (ICA) of functional magnetic resonance imaging (fMRI) data is commonly carried out under the assumption that each source may be represented as a spatially fixed pattern of activation, which leads to the instantaneous mixing model. To allow modeling patterns of spatio-temporal dynamics, in particular, the flow of oxygenated blood, we have developed a convolutive ICA approach: spatial complex ICA applied to frequency-domain fMRI data. In several frequency-bands, we identify components pertaining to activity in primary visual cortex (V1) and blood supply vessels. One such component, obtained in the 0.10 Hz band, is analyzed in detail and found to likely reflect flow of oxygenated blood in V1.

show abstract

A generalization of blind source separation algorithms for convolutive mixtures based on second-order statistics

Cited by 231 publications

References 27 publications

Convolutive Blind Source Separation Based on Disjointness Maximization of Subband Signals

Convolutive Blind Source Separation Based on Disjointness Maximization of Subband Signals

Variational EM for binaural sound-source separation and localization

Spatio-temporal dynamics in fMRI recordings revealed with complex independent component analysis

Contact Info

Product

Resources

About