Joint Detection and Estimation of Speech Spectral Amplitude Using Noncontinuous Gain Functions

Momeni, Hajar; Abutalebi, Hamid Reza; Tadaion, Aliakbar

doi:10.1109/taslp.2015.2427522

Cited by 16 publications

(15 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Such a combination would avoid any prior on the speech probability of presence since, by construction, the Neyman-Pearson test would not require such knowledge. A preliminary answer to this question is proposed in [30]. By exploiting [31,Theorem 1] to derive a detector that feeds an estimator based on a non-continuous gain function, this solution is a continuation and extension of [26].…”

Section: B State-of-the-artmentioning

confidence: 99%

“…Instead of having a prior structure for the estimator as in [30], the present paper addresses the problem of deriving both the detector and the estimator from a given estimation risk. We restrict our attention to the single-channel case.…”

Section: Contributionsmentioning

confidence: 99%

“…In a nutshell, we exploit [31,Theorem 2] to derive a theoretical framework for joint detection/estimation of speech signals in noise, where the detection is performed by a Neyman-Pearson test and the denoising by a Bayesian estimator. In contrast to [30], the derivation requires no prior knowledge on the estimator. We investigate three Bayesian estimation risks chosen according to the underlying binary hypothesis testing problems.…”

Section: Contributionsmentioning

confidence: 99%

“…The resulting framework yields three different and optimal joint detection-estimation algorithms that are assessed experimentally. The framework involves no psycho-physics and the resulting algorithms induce no performance loss in intelligibility and quality, and even bring some improvement in comparison to [29] and [30], with a reduced number of parameters.…”

Section: Contributionsmentioning

confidence: 99%

See 3 more Smart Citations

Optimal Bayesian Speech Enhancement by Parametric Joint Detection and Estimation

Pastor

Bey

2020

IEEE Access

View full text Add to dashboard Cite

In this paper, we propose a general framework to estimate short-time spectral amplitudes (STSA) of speech signals in noise by joint speech detection and estimation to remove or reduce background noise, without increasing signal distortion. The approach is motivated by the fact that speech signals have sparse time-frequency representations and can reasonably be assumed not to be present in every timefrequency bin of the time-frequency domain. By combining parametric detection and estimation theories, the main idea is to take into consideration speech presence and absence in each time-frequency bin to improve the performance of Bayesian estimators. In this respect, for three Bayesian estimators, optimal Neyman-Pearson detectors are derived to decide on the absence or presence of speech in each given timefrequency bin. Decisions returned by such detectors are then used to improve the initial estimates. The resulting estimations have been assessed in two scenarios, namely, with and without reference noise power spectrum. The objective tests confirm the relevance of these approaches, both in terms of speech quality and intelligibility. INDEX TERMS Unsupervised speech enhancement, parametric method, joint detection and estimation, Bayesian estimator, minimum mean square error (MMSE).

show abstract

Section: B State-of-the-artmentioning

confidence: 99%

Section: Contributionsmentioning

confidence: 99%

Section: Contributionsmentioning

confidence: 99%

Section: Contributionsmentioning

confidence: 99%

See 2 more Smart Citations

Optimal Bayesian Speech Enhancement by Parametric Joint Detection and Estimation

Pastor

Bey

2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Note that the gains of the STSA and the LSA estimator are the function of two variables ξ and γ (see (7) and (9)). The a priori SNR ξ is thus determined by decision-directed approach whose input is the noisy STSA R. As shown in Figure 1 the a posteriori SNR γ 1 is calculated from the rough enhanced STSA A 1 provided by SSBS.…”

Section: A Proposed Denoising Algorithmsmentioning

confidence: 99%

Joint Soft Threshold and Statistical Estimation for Speech Enhancement

Mai

Pastor

Bey

2018

2018 9th International Symposium on Signal, Image, Video and Communications (ISIVC)

View full text Add to dashboard Cite

This paper presents a novel method for speech enhancement based on the combination of sigmoid shrinkage and bayesian estimator. The main idea is to apply a joint detection and estimation to noisy speech before using a standard minimum-mean-squared-error (MMSE) estimator. Hence, the proposed method can take advantage of two basic approaches for improving the quality of noisy speech. Experiments performed on stationary and non-stationary noisy speech signals show that the proposed approach is promising when compared to classical methods, in terms of objective and pseudo-subjective measurements.

show abstract

Improvement of Power-Spectral-Subtraction Algorithm Using Cross-Term Compensation for Speech Enhancement

Chen

Shen

Wang

et al. 2017

Lecture Notes in Electrical Engineering

View full text Add to dashboard Cite

Joint Detection and Estimation of Speech Spectral Amplitude Using Noncontinuous Gain Functions

Cited by 16 publications

References 20 publications

Optimal Bayesian Speech Enhancement by Parametric Joint Detection and Estimation

Optimal Bayesian Speech Enhancement by Parametric Joint Detection and Estimation

Joint Soft Threshold and Statistical Estimation for Speech Enhancement

Improvement of Power-Spectral-Subtraction Algorithm Using Cross-Term Compensation for Speech Enhancement

Contact Info

Product

Resources

About