Hyper-parameterization of sparse reconstruction for speech enhancement

Shi, Yue; Low, Siow Yong; Yiu, Ka Fai Cedric

doi:10.1016/j.apacoust.2018.03.020

Cited by 9 publications

(6 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There have been many studies about regularization parameter selection criteria (see e.g. Karl 2005;Hansen 2000;Shi et al 2018) and as a future addition we anticipate that implementing an L-hypersurface criteria using the fixed-point optimization method from Belge et al (2002) would be advantageous. For all of these potential extensions to the existing framework, we note that current and future users can take advantage of cs-romer's object-oriented programming paradigm in order to facilitate the inclusion of enhancements in a straight forward manner.…”

Section: Discussionmentioning

confidence: 97%

“…Many attempts have been made to automatically calculate the regularization coefficients, 𝜂 1 and 𝜂 2 , (see e.g. Karl 2005;Hansen 2000;Belge et al 2002;Shi et al 2018). For this framework, and only for L1 regularization, we adopt the error bound calculation in Carrillo et al (2014); Pratley et al (2017).…”

Section: Cs-romermentioning

confidence: 99%

“…Following Shi et al (2018), df is an estimate of the degrees of freedom of 𝑥, which is defined as the number of non-zero entries.…”

Section: Model Selection Criteria: Aic and Bicmentioning

confidence: 99%

“…Although in Shi et al (2018) AIC and BIC have been used as a part of the optimisation process, here we calculate them using the estimated data only after convergence. In summary, the only difference between the AIC and BIC is that the BIC takes into account the logarithm of the total size of the estimator.…”

Section: Model Selection Criteria: Aic and Bicmentioning

confidence: 99%

See 3 more Smart Citations

CS-ROMER: A novel compressed sensing framework for Faraday depth reconstruction

Cárcamo,

Scaife,

Alexander

et al. 2022

Preprint

View full text Add to dashboard Cite

The reconstruction of Faraday depth structure from incomplete spectral polarization radio measurements using the RM Synthesis technique is an under-constrained problem requiring additional regularisation. In this paper we present cs-romer: a novel object-oriented compressed sensing framework to reconstruct Faraday depth signals from spectro-polarization radio data. Unlike previous compressed sensing applications, this framework is designed to work directly with data that are irregularly sampled in wavelength-squared space and to incorporate multiple forms of compressed sensing regularisation. We demonstrate the framework using simulated data for the JVLA telescope under a variety of observing conditions, and we introduce a methodology for identifying the optimal basis function for reconstruction of these data, using an approach that can also be applied to datasets from other telescopes and over different frequency ranges. In this work we show that the delta basis function provides optimal reconstruction for JVLA L-band data and we use this basis with observations of the low-mass galaxy cluster Abell 1314 in order to reconstruct the Faraday depth of its constituent cluster galaxies. We use the cs-romer framework to de-rotate the Galactic Faraday depth contribution directly from the wavelength-squared data and to handle the spectral behaviour of different radio sources in a direction-dependent manner. The results of this analysis show that individual galaxies within Abell 1314 deviate from the behaviour expected for a Faraday-thin screen such as the intra-cluster medium and instead suggest that the Faraday rotation exhibited by these galaxies is dominated by their local environments.

show abstract

Section: Discussionmentioning

confidence: 97%

Section: Cs-romermentioning

confidence: 99%

“…Following Shi et al (2018), df is an estimate of the degrees of freedom of 𝑥, which is defined as the number of non-zero entries.…”

Section: Model Selection Criteria: Aic and Bicmentioning

confidence: 99%

Section: Model Selection Criteria: Aic and Bicmentioning

confidence: 99%

See 2 more Smart Citations

CS-ROMER: A novel compressed sensing framework for Faraday depth reconstruction

Cárcamo,

Scaife,

Alexander

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…As the sparsest solution may not necessarily translate to an improvement in the overall speech quality, the sparsity level of the output should match closely to the desired signal [23]. In fact, as far as speech enhancement applications are concerned, the sparsity level can be viewed as the noise suppression level.…”

mentioning

confidence: 99%

Two-level optimization approach with accelerated proximal gradient for objective measures in sparse speech reconstruction

Dam

Low

Nordholm

2022

JIMO

Self Cite

View full text Add to dashboard Cite

<p style='text-indent:20px;'>Compressive speech enhancement makes use of the sparseness of speech and the non-sparseness of noise in time-frequency representation to perform speech enhancement. However, reconstructing the sparsest output may not necessarily translate to a good enhanced speech signal as speech distortion may be at risk. This paper proposes a two level optimization approach to incorporate objective quality measures in compressive speech enhancement. The proposed method combines the accelerated proximal gradient approach and a global one dimensional optimization method to solve the sparse reconstruction. By incorporating objective quality measures in the optimization process, the reconstructed output is not only sparse but also maintains the highest objective quality score possible. In other words, the sparse speech reconstruction process is now quality sparse speech reconstruction. Experimental results in a compressive speech enhancement consistently show score improvement in objectives measures in different noisy environments compared to the non-optimized method. Additionally, the proposed optimization yields a higher convergence rate with a lower computational complexity compared to the existing methods.</p>

show abstract

Evaluation of speech enhancement algorithms applied to electrolaryngeal speech degraded by noise

Mathew¹,

Gopakumar²

2021

Applied Acoustics

View full text Add to dashboard Cite

Hyper-parameterization of sparse reconstruction for speech enhancement

Cited by 9 publications

References 37 publications

CS-ROMER: A novel compressed sensing framework for Faraday depth reconstruction

CS-ROMER: A novel compressed sensing framework for Faraday depth reconstruction

Two-level optimization approach with accelerated proximal gradient for objective measures in sparse speech reconstruction

Evaluation of speech enhancement algorithms applied to electrolaryngeal speech degraded by noise

Contact Info

Product

Resources

About