Sparse target cancellation filters with application to semi-blind noise extraction

Málek, Jiřı́; Koldovsky, Zbynek

doi:10.1109/icassp.2014.6853971

Cited by 9 publications

(4 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, neither formulation ( 21) nor (22) takes into account the fact that f contains certain estimation errors. It is therefore better to relax the constraint given through (20). One such alternative to (22) is LASSO (Least Absolute Shrinkage and Selector Operator) defined as…”

Section: Sparse Solutions Of (20)mentioning

confidence: 99%

“…A novel strategy is used in [18], [19], [20] by considering the fact that relative impulse responses can be replaced or approximated by sparse filters, that is, by filters that have many coefficients equal to zero; see also [21], a recent work on sparse approximations of room impulse responses. The authors of [20] propose a semi-blind approach assuming knowledge of the support of a sparse approximation. Hence only nonzero coefficients are estimated using ICA, which implies a significant dimensionality reduction of the parameter space.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Spatial Source Subtraction Based on Incomplete Measurements of Relative Transfer Function

Koldovsky

Málek

Gannot

2015

IEEE/ACM Trans. Audio Speech Lang. Process.

Self Cite

View full text Add to dashboard Cite

Relative impulse responses between microphones are usually long and dense due to the reverberant acoustic environment. Estimating them from short and noisy recordings poses a long-standing challenge of audio signal processing. In this paper we apply a novel strategy based on ideas of Compressed Sensing. Relative transfer function (RTF) corresponding to the relative impulse response can often be estimated accurately from noisy data but only for certain frequencies. This means that often only an incomplete measurement of the RTF is available. A complete RTF estimate can be obtained through finding its sparsest representation in the time-domain: that is, through computing the sparsest among the corresponding relative impulse responses. Based on this approach, we propose to estimate the RTF from noisy data in three steps. First, the RTF is estimated using any conventional method such as the non-stationarity-based estimator by Gannot et al. or through Blind Source Separation. Second, frequencies are determined for which the RTF estimate appears to be accurate. Third, the RTF is reconstructed through solving a weighted 1 convex program, which we propose to solve via a computationally efficient variant of the SpaRSA (Sparse Reconstruction by Separable Approximation) algorithm. An extensive experimental study with real-world recordings has been conducted. It has been shown that the proposed method is capable of improving many conventional estimators used as the first step in most situations.

show abstract

Section: Sparse Solutions Of (20)mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Spatial Source Subtraction Based on Incomplete Measurements of Relative Transfer Function

Koldovsky

Málek

Gannot

2015

IEEE/ACM Trans. Audio Speech Lang. Process.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Both types of methods estimate an RTF for each channel pair separately and are suited only for the case of a single desired signal. Other approaches include independent component analysis, e.g., [5] which is applicable in underdetermined scenarios, and [6] that uses sparsity.…”

Section: Introductionmentioning

confidence: 99%

Estimation of relative transfer function in the presence of stationary noise based on segmental power spectral density matrix subtraction

Girin

Horaud

et al. 2015

2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

This paper addresses the problem of relative transfer function (RTF) estimation in the presence of stationary noise. We propose an RTF identification method based on segmental power spectral density (PSD) matrix subtraction. First multiple channel microphone signals are divided into segments corresponding to speech-plus-noise activity and noise-only. Then, the subtraction of two segmental PSD matrices leads to an almost noise-free PSD matrix by reducing the stationary noise component and preserving non-stationary speech component. This noise-free PSD matrix is used for single speaker RTF identification by eigenvalue decomposition. Experiments are performed in the context of sound source localization to evaluate the efficiency of the proposed method.

show abstract

“…To solve, the fact that ReIRs are fast decaying (compressible) sequences is used. Such sequences, al-though being long and dense, can be efficiently approximated by sparse surrogates; see, e.g., [10]. Convex programming based on 1 -norm minimization is used to find the optimum sparse representation.…”

Section: Introductionmentioning

confidence: 99%

Sparse reconstruction of incomplete relative transfer function: Discrete and continuous time domain

Koldovsky

Tichavský

2015

2015 23rd European Signal Processing Conference (EUSIPCO)

Self Cite

View full text Add to dashboard Cite

Relative transfer functions (RTF) between microphones can often be estimated accurately but only for certain frequencies. For example, this happens in situations where the RTF is estimated from a noise-free signal of a target source whose spectrum does not span the whole frequency range. By combining a conventional RTF estimator and a selection of the active frequencies, an incomplete measurement of the RTF is obtained. We propose to retrieve the whole RTF estimate through finding the sparsest representation of the incomplete measurement in the discrete or continuous time-domain and compare both approaches. The RTF estimates are evaluated in terms of attenuation rate that measures the target signal cancellation at the output of a blocking matrix. It is shown by experiments that the reconstructed estimate can achieve significantly better attenuation than the initial (complete) estimate.

show abstract

Sparse target cancellation filters with application to semi-blind noise extraction

Cited by 9 publications

References 23 publications

Spatial Source Subtraction Based on Incomplete Measurements of Relative Transfer Function

Spatial Source Subtraction Based on Incomplete Measurements of Relative Transfer Function

Estimation of relative transfer function in the presence of stationary noise based on segmental power spectral density matrix subtraction

Sparse reconstruction of incomplete relative transfer function: Discrete and continuous time domain

Contact Info

Product

Resources

About