A machine learning based speech enhancement method is proposed to improve the intelligibility of whispered speech. A binary mask estimated by a two-class support vector machine （SVM） classifier is used to synthesize...A machine learning based speech enhancement method is proposed to improve the intelligibility of whispered speech. A binary mask estimated by a two-class support vector machine （SVM） classifier is used to synthesize the enhanced whisper. A novel noise robust feature called Gammatone feature cosine coefficients （GFCCs） extracted by an auditory periphery model is derived and used for the binary mask estimation. The intelligibility performance of the proposed method is evaluated and compared with the traditional speech enhancement methods. Objective and subjective evaluation results indicate that the proposed method can effectively improve the intelligibility of whispered speech which is contaminated by noise. Compared with the power subtract algorithm and the log-MMSE algorithm, both of which do not improve the intelligibility in lower signal-to-noise ratio （SNR） environments, the proposed method has good performance in improving the intelligibility of noisy whisper. Additionally, the intelligibility of the enhanced whispered speech using the proposed method also outperforms that of the corresponding unprocessed noisy whispered speech.展开更多
State estimation is the precondition and foundation of a bioprocess monitoring and optimal control. However,there are many difficulties in dealing with a non-linear system,such as the instability of process, un-modele...State estimation is the precondition and foundation of a bioprocess monitoring and optimal control. However,there are many difficulties in dealing with a non-linear system,such as the instability of process, un-modeled dynamics,parameter sensitivity,etc.This paper discusses the principles and characteristics of three different approaches,extended Kalman filters,strong tracking filters and unscented transformation based Kalman filters.By introducing the unscented transformation method and a sub-optimal fading factor to correct the prediction error covariance,an improved Kalman filter,unscented transformation based robust Kalman filter,is proposed. The performance of the algorithm is compared with the strong tracking filter and unscented transformation based Kalman filter and illustrated in a typical case study for glutathione fermentation process.The results show that the proposed algorithm presents better accuracy and stability on the state estimation in numerical calculations.展开更多
In this paper,a reinforced gradient-type iterative learning control pro file is proposed by making use of system matrices and a proper learning step to improve the tracking performance of batch processes disturbed by ...In this paper,a reinforced gradient-type iterative learning control pro file is proposed by making use of system matrices and a proper learning step to improve the tracking performance of batch processes disturbed by external Gaussian white noise.The robustness is analyzed and the range of the step is speci fied by means of statistical technique and matrix theory.Compared with the conventional one,the proposed algorithm is more ef ficient to resist external noise.Numerical simulations of an injection molding process illustrate that the proposed scheme is feasible and effective.展开更多
This paper considers experimental situations where the interested effects have to be or- thogonal to a set of nonnegligible effects. It is shown that various types of orthogonal arrays with mixed strength are A-optima...This paper considers experimental situations where the interested effects have to be or- thogonal to a set of nonnegligible effects. It is shown that various types of orthogonal arrays with mixed strength are A-optimal for estimating the parameters in ANOVA high dimension model representation. Both cases including interactions or not are considered in the model. In particularly, the estimations of all main effects are A-optimal in a mixed strength （2, 2）3 orthogonal array and the estimations of all main effects and two-factor interactions in G~ x G~ are A-optimal in a mixed strength （2, 2）4 orthogonal array. The properties are also illustrated through a simulation study.展开更多
As a competitive depth, L^2-depth is modified from L^2-depth. Its induced median is called L^2-median. Basic properties of the median and its sample version are provided. Especially, the strong consistency of sample m...As a competitive depth, L^2-depth is modified from L^2-depth. Its induced median is called L^2-median. Basic properties of the median and its sample version are provided. Especially, the strong consistency of sample median is gained under weaker condition. Robustness of the median and its sample version is discussed. Besides ease of computation, it is shown that L^2-median has both good large-sample and robust properties. Simulation studies are also given to compare the breakdown point of L^2-median with that of other depth-induced medians.展开更多
基金The National Natural Science Foundation of China (No.61231002,61273266,51075068,60872073,60975017, 61003131)the Ph.D.Programs Foundation of the Ministry of Education of China(No.20110092130004)+1 种基金the Science Foundation for Young Talents in the Educational Committee of Anhui Province(No. 2010SQRL018)the 211 Project of Anhui University(No.2009QN027B)
文摘A machine learning based speech enhancement method is proposed to improve the intelligibility of whispered speech. A binary mask estimated by a two-class support vector machine （SVM） classifier is used to synthesize the enhanced whisper. A novel noise robust feature called Gammatone feature cosine coefficients （GFCCs） extracted by an auditory periphery model is derived and used for the binary mask estimation. The intelligibility performance of the proposed method is evaluated and compared with the traditional speech enhancement methods. Objective and subjective evaluation results indicate that the proposed method can effectively improve the intelligibility of whispered speech which is contaminated by noise. Compared with the power subtract algorithm and the log-MMSE algorithm, both of which do not improve the intelligibility in lower signal-to-noise ratio （SNR） environments, the proposed method has good performance in improving the intelligibility of noisy whisper. Additionally, the intelligibility of the enhanced whispered speech using the proposed method also outperforms that of the corresponding unprocessed noisy whispered speech.
基金Supported by the National Natural Science Foundation of China （20476007, 20676013）.
文摘State estimation is the precondition and foundation of a bioprocess monitoring and optimal control. However,there are many difficulties in dealing with a non-linear system,such as the instability of process, un-modeled dynamics,parameter sensitivity,etc.This paper discusses the principles and characteristics of three different approaches,extended Kalman filters,strong tracking filters and unscented transformation based Kalman filters.By introducing the unscented transformation method and a sub-optimal fading factor to correct the prediction error covariance,an improved Kalman filter,unscented transformation based robust Kalman filter,is proposed. The performance of the algorithm is compared with the strong tracking filter and unscented transformation based Kalman filter and illustrated in a typical case study for glutathione fermentation process.The results show that the proposed algorithm presents better accuracy and stability on the state estimation in numerical calculations.
基金Supported by National Natural Science Foundation of China(F010114-6097414061273135)
文摘In this paper,a reinforced gradient-type iterative learning control pro file is proposed by making use of system matrices and a proper learning step to improve the tracking performance of batch processes disturbed by external Gaussian white noise.The robustness is analyzed and the range of the step is speci fied by means of statistical technique and matrix theory.Compared with the conventional one,the proposed algorithm is more ef ficient to resist external noise.Numerical simulations of an injection molding process illustrate that the proposed scheme is feasible and effective.
基金the National Natural Science Foundation of China under Grant Nos.11171065,11301073the Natural Science Foundation of Jiangsu under Grant No.BK20141326+1 种基金the Research Fund for the Doctoral Program of Higher Education of China under Grant No.20120092110021Scientific Research Foundation of Graduate School of Southeast University under Grant No.YBJJ1444
文摘This paper considers experimental situations where the interested effects have to be or- thogonal to a set of nonnegligible effects. It is shown that various types of orthogonal arrays with mixed strength are A-optimal for estimating the parameters in ANOVA high dimension model representation. Both cases including interactions or not are considered in the model. In particularly, the estimations of all main effects are A-optimal in a mixed strength （2, 2）3 orthogonal array and the estimations of all main effects and two-factor interactions in G~ x G~ are A-optimal in a mixed strength （2, 2）4 orthogonal array. The properties are also illustrated through a simulation study.
基金This research is supported by the National Natural Science Foundation of China under Grant No. 10971007, the Specialized Research Fund for the Doctoral Program of Higher Education under Grant No. 20091103120012, and the research fund of B JUT under Grant No. X0006013200904. The authors greatly appreciate the constructive comments by the three anonymous reviewers and an associate editor. This has led to substantial improvements in our paper. Special thanks to financial support from the Science and Technology Development Fund of the Government of the Macao Special Administrative Region （No. 045/2005/A）.
文摘As a competitive depth, L^2-depth is modified from L^2-depth. Its induced median is called L^2-median. Basic properties of the median and its sample version are provided. Especially, the strong consistency of sample median is gained under weaker condition. Robustness of the median and its sample version is discussed. Besides ease of computation, it is shown that L^2-median has both good large-sample and robust properties. Simulation studies are also given to compare the breakdown point of L^2-median with that of other depth-induced medians.
