Sparse Coding for Convolutive Blind Audio Source Separation

Jafari, Maria G.; Abdallah, Samer A.; Plumbley, Mark D.; Davies, Mike E.

doi:10.1007/11679363_17

Cited by 14 publications

(11 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In an earlier preliminary investigation [33], we found that the objective SDR measures did not always correspond to our perceived quality of the separation. This difference may be due to the calculation of the objective criteria requires a reconstruction filter to be estimated, which is non-trivial for convolutive mixing or to distortions which are perceptually minor but which are not allowed for by the (linear, time-invariant) filter [34].…”

Section: Objective Evaluationmentioning

confidence: 53%

“…For the DUET algorithm we used an STFT frame size of 1024 samples, which was found by Yilmaz and Rickard [11] to give the best separation performance at 16 kHz. For the proposed adaptive stereo basis algorithm, we used an adaptive basis frame size of 512 samples, to be consistent with preliminary experiments which indicated that this would be sufficient for separation at a 16 kHz sampling rate with reasonable room reverberation times [33]. Excerpts of the original mixture and source signals and of the estimated source signals are available for listening on our demo web page 2 .…”

Section: Methodsmentioning

confidence: 94%

“…We used the FD-ICA algorithm with the MuSIC-based permutation alignment algorithm described by Mitianoudis and Davies [9], setting the STFT frame size to 2048 samples, which was previously found to be appropriate for this algorithm at a 16kHz sampling rate [9,33]. For the DUET algorithm we used an STFT frame size of 1024 samples, which was found by Yilmaz and Rickard [11] to give the best separation performance at 16 kHz.…”

Section: Methodsmentioning

confidence: 99%

See 2 more Smart Citations

An adaptive stereo basis method for convolutive blind audio source separation

et al. 2008

Self Cite

View full text Add to dashboard Cite

We consider the problem of convolutive blind source separation of stereo mixtures, where a pair of microphones records mixtures of sound sources that are convolved with the impulse response between each source and sensor. We propose an Adaptive Stereo Basis (ASB) source separation method for such convolutive mixtures, using an adaptive transform basis which is learned from the stereo mixture pair.The stereo basis vector pairs of the transform are grouped according to the estimated relative delay between the left and right channels for each basis, and the sources are then extracted by projecting the transformed signal onto the subspace corresponding to each group of basis vector pairs. The performance of the proposed algorithm is compared with FD-ICA and DUET under different reverberation and noise conditions, using both objective distortion measures and formal listening tests. Preprint submitted to Elsevier Science 10 August 2007NOTICE: this is the author's version of a work that was accepted for publication in Neurocomputing. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in : Neurocomputing 71(10-12), 2087-2097, June 2008. doi: 10.1016/j.neucom.2007 The results indicate that the proposed stereo coding method is competitive with both these algorithms at short and intermediate reverberation times, and offers significantly improved performance at low noise and short reverberation times.

show abstract

Section: Objective Evaluationmentioning

confidence: 53%

Section: Methodsmentioning

confidence: 94%

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

An adaptive stereo basis method for convolutive blind audio source separation

et al. 2008

Self Cite

View full text Add to dashboard Cite

show abstract

“…The key differences between different SCA techniques boil down to the method of clustering the components for mixing matrix estimation as well as mask construction or sparse recovery to achieve source separation (Jafari et al, 2006;Mourad and Reilly, 2010). Generalized to CASA, in many SCA approaches, a soft mask is applied thus the assumption that each spectro-temporal coefficient belongs to the same source is relaxed and the recovered speech does not suffer from the musical noise and missing components in their spectrographic representation (Araki et al, 2005;Kearns et al, 1997).…”

Section: Prior Workmentioning

confidence: 99%

Computational methods for underdetermined convolutive speech localization and separation via model-based sparse component analysis

Asaei

Bourlard

Taghizadeh

et al. 2016

Speech Communication

View full text Add to dashboard Cite

In this paper, the problem of speech source localization and separation from recordings of convolutive underdetermined mixtures is studied. The problem is cast as recovering the spatio-spectral speech information embedded in a microphone array compressed measurements of the acoustic field. A model-based sparse component analysis framework is formulated for sparse reconstruction of the speech spectra in a reverberant acoustic resulting in joint localization and separation of the individual sources. We compare and contrast the computational approaches to model-based sparse recovery exploiting spatial sparsity as well as spectral structures underlying spectrographic representation of speech signals. In this context, we explore identification of the sparsity structures at the auditory and acoustic representation spaces. The auditory structures are formulated upon the principles of structural grouping based on proximity, autoregressive correlation and harmonicity of the spectral coefficients and they are incorporated for sparse reconstruction. The acoustic structures are formulated upon the image model of multipath propagation and they are exploited to characterize the compressive measurement matrix associated with microphone array recordings.Three approaches to sparse recovery relying on combinatorial optimization, convex relaxation and Bayesian methods are studied and evaluated based on thorough experiments. The sparse Bayesian learning method is shown to yield better perceptual quality while the interference suppression is also achieved using the combinatorial approach with the advantage of offering the most efficient computational cost. Furthermore, it is demonstrated that an average autoregressive model can be learned for speech localization and exploiting the proximity structure in the form of block sparse coefficients enables accurate localization. Throughout the extensive empirical evaluation, we confirm that a large and random placement of the microphones enables significant improvement in source localization and separation performance.

show abstract

“…It is often difficult to determine a relationship between a class of signals and a particular dictionary, especially for natural signals [6]. This has led researchers to look for learned, rather than fixed dictionaries, using techniques such as independent component analysis (ICA), as the underlying learning algorithm [2], [7], [8]. These methods, however, are computationally very expensive, and often require fairly large datasets in order to learn the dictionary bases.…”

Section: Introductionmentioning

confidence: 99%

An adaptive orthogonal sparsifying transform for speech signals

Jafari

Plumbley

2008

2008 3rd International Symposium on Communications, Control and Signal Processing

Self Cite

View full text Add to dashboard Cite

In this paper we consider the problem of representing a speech signal with an adaptive transform that captures the main features of the data. The transform is orthogonal by construction, and is found to give a sparse representation of the data being analysed. The orthogonality property implies that evaluation of both the forward and inverse transform involve a simple matrix multiplication. The proposed dictionary learning algorithm is compared to the K singular value decomposition (K-SVD) method, which is found to yield very sparse representations, at the cost of a high approximation error. The proposed algorithm is shown to have a much lower computational complexity than K-SVD, while the resulting signal representation remains relatively sparse.

show abstract

Sparse Coding for Convolutive Blind Audio Source Separation

Cited by 14 publications

References 9 publications

An adaptive stereo basis method for convolutive blind audio source separation

An adaptive stereo basis method for convolutive blind audio source separation

Computational methods for underdetermined convolutive speech localization and separation via model-based sparse component analysis

An adaptive orthogonal sparsifying transform for speech signals

Contact Info

Product

Resources

About