Enhanced speech based on the traditional wavelet threshold function had auditory oscillation distortion and the low signal-to-noise ratio (SNR). In order to solve these problems, a new continuous differentiable thresh...Enhanced speech based on the traditional wavelet threshold function had auditory oscillation distortion and the low signal-to-noise ratio (SNR). In order to solve these problems, a new continuous differentiable threshold function for speech enhancement was presented. Firstly, the function adopted narrow threshold areas, preserved the smaller signal speech, and improved the speech quality; secondly, based on the properties of the continuous differentiable and non-fixed deviation, each area function was attained gradually by using the method of mathematical derivation. It ensured that enhanced speech was continuous and smooth; it removed the auditory oscillation distortion; finally, combined with the Bark wavelet packets, it further improved human auditory perception. Experimental results show that the segmental SNR and PESQ (perceptual evaluation of speech quality) of the enhanced speech using this method increase effectively, compared with the existing speech enhancement algorithms based on wavelet threshold.展开更多
In non-cooperation communication, speech suppression is an important link of aircraft type recognition based on shortwave speech communication. Ensemble mean and ensemble variance for analyzing physical characteristic...In non-cooperation communication, speech suppression is an important link of aircraft type recognition based on shortwave speech communication. Ensemble mean and ensemble variance for analyzing physical characteristics of aircraft cockpit background sound has been proposed in this paper. Some important peak feature is revealed from the ensemble mean and ensemble variance, which has important guiding significance to further researches on speech suppression of aircraft type recognition based on shortwave speech communication. The two algorithms of speech suppression are proposed which are empirical mode decomposition and wavelet transform(EMD_WT) and the algorithm of ensemble empirical mode decomposition and wavelet packet (EEMD_WP). In the process of speech suppression, EMD_WT produced extra noise pollution, and EEMD_WP kept the aircraft cockpit background sound and weakened the effect of the speech largely. The contrast test proves that EEMD_WP has better performance by in time domain and in frequency domain.展开更多
Support vector machine(SVM)has a good application prospect for speech recognition problems;still optimum parameter selection is a vital issue for it.To improve the learning ability of SVM,a method for searching the op...Support vector machine(SVM)has a good application prospect for speech recognition problems;still optimum parameter selection is a vital issue for it.To improve the learning ability of SVM,a method for searching the optimal parameters based on integration of predator prey optimization(PPO)and Hooke-Jeeves method has been proposed.In PPO technique,population consists of prey and predator particles.The prey particles search the optimum solution and predator always attacks the global best prey particle.The solution obtained by PPO is further improved by applying Hooke-Jeeves method.Proposed method is applied to recognize isolated words in a Hindi speech database and also to recognize words in a benchmark database TI-20 in clean and noisy environment.A recognition rate of 81.5%for Hindi database and 92.2%for TI-20 database has been achieved using proposed technique.展开更多
Wavelet packets decompose signals in to broader components using linear spectral bisecting. Mixing matrix is the key issue in the Blind Source Separation (BSS) literature especially in under-determined cases. In this ...Wavelet packets decompose signals in to broader components using linear spectral bisecting. Mixing matrix is the key issue in the Blind Source Separation (BSS) literature especially in under-determined cases. In this paper, we propose a simple and novel method in Short Time Wavelet Packet (STWP) analysis to estimate blindly the mixing matrix of speech signals from noise free linear mixtures in over-complete cases. In this paper, the Laplacian model is considered in short time-wavelet packets and is applied to each histogram of packets. Expectation Maximization (EM) algorithm is used to train the model and calculate the model parameters. In our simulations, comparison with the other recent results will be computed and it is shown that our results are better than others. It is shown that complexity of computation of model is decreased and consequently the speed of convergence is increased.展开更多
针对目前IP电话语音质量难以准确评价及测量的情况,研究了一种基于E-Model的VoIP(voice over internet protocol)语音质量的测量模型。该模型考虑了IP网络中大多数的网络损伤因素,并能容易地计算出不同丢包率、不同的延迟和抖动所对应的...针对目前IP电话语音质量难以准确评价及测量的情况,研究了一种基于E-Model的VoIP(voice over internet protocol)语音质量的测量模型。该模型考虑了IP网络中大多数的网络损伤因素,并能容易地计算出不同丢包率、不同的延迟和抖动所对应的MOS(mean opinion score)值。测试出IP电话在网络中质量变化情况,有利于IP网络中资源的调整和VoIP质量的提高。展开更多
基金Project(61072087) supported by the National Natural Science Foundation of ChinaProject(2011-035) supported by Shanxi Province Scholarship Foundation, China+2 种基金Project(20120010) supported by Universities High-tech Foundation Projects, ChinaProject (2013021016-1) supported by the Youth Science and Technology Foundation of Shanxi Province, ChinaProjects(2013011016-1, 2012011014-1) supported by the Natural Science Foundation of Shanxi Province, China
文摘Enhanced speech based on the traditional wavelet threshold function had auditory oscillation distortion and the low signal-to-noise ratio (SNR). In order to solve these problems, a new continuous differentiable threshold function for speech enhancement was presented. Firstly, the function adopted narrow threshold areas, preserved the smaller signal speech, and improved the speech quality; secondly, based on the properties of the continuous differentiable and non-fixed deviation, each area function was attained gradually by using the method of mathematical derivation. It ensured that enhanced speech was continuous and smooth; it removed the auditory oscillation distortion; finally, combined with the Bark wavelet packets, it further improved human auditory perception. Experimental results show that the segmental SNR and PESQ (perceptual evaluation of speech quality) of the enhanced speech using this method increase effectively, compared with the existing speech enhancement algorithms based on wavelet threshold.
基金Sponsored by the National Natural Science Foundation of China(Grant No.60975019)the National High Technology Research and Development Program of China(Grant No.2009AA04Z215)
文摘In non-cooperation communication, speech suppression is an important link of aircraft type recognition based on shortwave speech communication. Ensemble mean and ensemble variance for analyzing physical characteristics of aircraft cockpit background sound has been proposed in this paper. Some important peak feature is revealed from the ensemble mean and ensemble variance, which has important guiding significance to further researches on speech suppression of aircraft type recognition based on shortwave speech communication. The two algorithms of speech suppression are proposed which are empirical mode decomposition and wavelet transform(EMD_WT) and the algorithm of ensemble empirical mode decomposition and wavelet packet (EEMD_WP). In the process of speech suppression, EMD_WT produced extra noise pollution, and EEMD_WP kept the aircraft cockpit background sound and weakened the effect of the speech largely. The contrast test proves that EEMD_WP has better performance by in time domain and in frequency domain.
文摘Support vector machine(SVM)has a good application prospect for speech recognition problems;still optimum parameter selection is a vital issue for it.To improve the learning ability of SVM,a method for searching the optimal parameters based on integration of predator prey optimization(PPO)and Hooke-Jeeves method has been proposed.In PPO technique,population consists of prey and predator particles.The prey particles search the optimum solution and predator always attacks the global best prey particle.The solution obtained by PPO is further improved by applying Hooke-Jeeves method.Proposed method is applied to recognize isolated words in a Hindi speech database and also to recognize words in a benchmark database TI-20 in clean and noisy environment.A recognition rate of 81.5%for Hindi database and 92.2%for TI-20 database has been achieved using proposed technique.
文摘Wavelet packets decompose signals in to broader components using linear spectral bisecting. Mixing matrix is the key issue in the Blind Source Separation (BSS) literature especially in under-determined cases. In this paper, we propose a simple and novel method in Short Time Wavelet Packet (STWP) analysis to estimate blindly the mixing matrix of speech signals from noise free linear mixtures in over-complete cases. In this paper, the Laplacian model is considered in short time-wavelet packets and is applied to each histogram of packets. Expectation Maximization (EM) algorithm is used to train the model and calculate the model parameters. In our simulations, comparison with the other recent results will be computed and it is shown that our results are better than others. It is shown that complexity of computation of model is decreased and consequently the speed of convergence is increased.
文摘针对目前IP电话语音质量难以准确评价及测量的情况,研究了一种基于E-Model的VoIP(voice over internet protocol)语音质量的测量模型。该模型考虑了IP网络中大多数的网络损伤因素,并能容易地计算出不同丢包率、不同的延迟和抖动所对应的MOS(mean opinion score)值。测试出IP电话在网络中质量变化情况,有利于IP网络中资源的调整和VoIP质量的提高。