Low-Distortion MMSE Speech Enhancement Estimator Based on Laplacian Prior
The most well-known conventional speech enhancement algorithms introduce unwanted artifact noise and speech distortion to the enhanced signal. Reducing the effects of such issues require more robust linear and non-linear estimators. This paper proposes new optimum linear and non-linear Laplacian dis...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2017-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/7914629/ |
| _version_ | 1818645006104133632 |
|---|---|
| author | Basheera M. Mahmmod Abd Rahman Ramli Sadiq H. Abdulhussian Syed Abdul Rahman Al-Haddad Wissam A. Jassim |
| author_facet | Basheera M. Mahmmod Abd Rahman Ramli Sadiq H. Abdulhussian Syed Abdul Rahman Al-Haddad Wissam A. Jassim |
| author_sort | Basheera M. Mahmmod |
| collection | DOAJ |
| description | The most well-known conventional speech enhancement algorithms introduce unwanted artifact noise and speech distortion to the enhanced signal. Reducing the effects of such issues require more robust linear and non-linear estimators. This paper proposes new optimum linear and non-linear Laplacian distribution-based estimators. The proposed estimators are derived based on a minimum mean squared error (MMSE) sense to minimize the distortion in different conditions of the underlying speech. Thus, artifact noise is reduced without compromising the noise reduction process. The analytical solutions of the Laplacian distribution-based estimators, linear bilateral Laplacian gain estimator (LBLG), and non-linear bilateral Laplacian gain estimator (NBLG), are presented. The proposed estimators are implemented in three steps. First, the observation signal is decorrelated through a real transform domain to obtain its transform coefficients. Second, the proposed estimators are applied to estimate the clean speech signal from the noisy signal in the decorrelated domain. Finally, the inverse of the real transform is applied to obtain the original speech signal in the time domain. Two conditions in these estimators account for interference events between the speech signal and noise coefficients in the decorrelated domain. Moreover, a mathematical aspect of mean square error of LBLG is evaluated, which presents a significant improvement over other methods. Furthermore, a comprehensive description of the whole variations of the LBLG and NBLG gains characteristics is presented. A comparative evaluation is performed with effective quality metrics, segmental signal-to-noise ratio and perceptual evaluation of speech quality, to demonstrate the advantage and effectiveness of the proposed estimators. The performance of the proposed estimators outperformed other methods, which are the traditional MMSE approach, perceptually motivated Bayesian estimator, dual gain Wiener estimator, and dual MMSE estimator in terms of different objective measurements. |
| first_indexed | 2024-12-17T00:23:52Z |
| format | Article |
| id | doaj.art-6e30ebff4427435a93b22bbfc36ab596 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-12-17T00:23:52Z |
| publishDate | 2017-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-6e30ebff4427435a93b22bbfc36ab5962022-12-21T22:10:30ZengIEEEIEEE Access2169-35362017-01-0159866988110.1109/ACCESS.2017.26997827914629Low-Distortion MMSE Speech Enhancement Estimator Based on Laplacian PriorBasheera M. Mahmmod0Abd Rahman Ramli1Sadiq H. Abdulhussian2https://orcid.org/0000-0002-6439-0082Syed Abdul Rahman Al-Haddad3Wissam A. Jassim4Department of Computer and Communication Systems Engineering, Universiti Putra Malaysia, Selangor, MalaysiaDepartment of Computer and Communication Systems Engineering, Universiti Putra Malaysia, Selangor, MalaysiaDepartment of Computer and Communication Systems Engineering, Universiti Putra Malaysia, Selangor, MalaysiaDepartment of Computer and Communication Systems Engineering, Universiti Putra Malaysia, Selangor, MalaysiaADAPT Center, School of Engineering, Trinity College Dublin, University of Dublin, Dublin, IrelandThe most well-known conventional speech enhancement algorithms introduce unwanted artifact noise and speech distortion to the enhanced signal. Reducing the effects of such issues require more robust linear and non-linear estimators. This paper proposes new optimum linear and non-linear Laplacian distribution-based estimators. The proposed estimators are derived based on a minimum mean squared error (MMSE) sense to minimize the distortion in different conditions of the underlying speech. Thus, artifact noise is reduced without compromising the noise reduction process. The analytical solutions of the Laplacian distribution-based estimators, linear bilateral Laplacian gain estimator (LBLG), and non-linear bilateral Laplacian gain estimator (NBLG), are presented. The proposed estimators are implemented in three steps. First, the observation signal is decorrelated through a real transform domain to obtain its transform coefficients. Second, the proposed estimators are applied to estimate the clean speech signal from the noisy signal in the decorrelated domain. Finally, the inverse of the real transform is applied to obtain the original speech signal in the time domain. Two conditions in these estimators account for interference events between the speech signal and noise coefficients in the decorrelated domain. Moreover, a mathematical aspect of mean square error of LBLG is evaluated, which presents a significant improvement over other methods. Furthermore, a comprehensive description of the whole variations of the LBLG and NBLG gains characteristics is presented. A comparative evaluation is performed with effective quality metrics, segmental signal-to-noise ratio and perceptual evaluation of speech quality, to demonstrate the advantage and effectiveness of the proposed estimators. The performance of the proposed estimators outperformed other methods, which are the traditional MMSE approach, perceptually motivated Bayesian estimator, dual gain Wiener estimator, and dual MMSE estimator in terms of different objective measurements.https://ieeexplore.ieee.org/document/7914629/Speech enhancementLaplacian distributionminimum mean square error estimatorspeech modelnoise model |
| spellingShingle | Basheera M. Mahmmod Abd Rahman Ramli Sadiq H. Abdulhussian Syed Abdul Rahman Al-Haddad Wissam A. Jassim Low-Distortion MMSE Speech Enhancement Estimator Based on Laplacian Prior IEEE Access Speech enhancement Laplacian distribution minimum mean square error estimator speech model noise model |
| title | Low-Distortion MMSE Speech Enhancement Estimator Based on Laplacian Prior |
| title_full | Low-Distortion MMSE Speech Enhancement Estimator Based on Laplacian Prior |
| title_fullStr | Low-Distortion MMSE Speech Enhancement Estimator Based on Laplacian Prior |
| title_full_unstemmed | Low-Distortion MMSE Speech Enhancement Estimator Based on Laplacian Prior |
| title_short | Low-Distortion MMSE Speech Enhancement Estimator Based on Laplacian Prior |
| title_sort | low distortion mmse speech enhancement estimator based on laplacian prior |
| topic | Speech enhancement Laplacian distribution minimum mean square error estimator speech model noise model |
| url | https://ieeexplore.ieee.org/document/7914629/ |
| work_keys_str_mv | AT basheerammahmmod lowdistortionmmsespeechenhancementestimatorbasedonlaplacianprior AT abdrahmanramli lowdistortionmmsespeechenhancementestimatorbasedonlaplacianprior AT sadiqhabdulhussian lowdistortionmmsespeechenhancementestimatorbasedonlaplacianprior AT syedabdulrahmanalhaddad lowdistortionmmsespeechenhancementestimatorbasedonlaplacianprior AT wissamajassim lowdistortionmmsespeechenhancementestimatorbasedonlaplacianprior |