首页 | 本学科首页   官方微博 | 高级检索  
相似文献
 共查询到19条相似文献,搜索用时 296 毫秒
1.
A blind speech source separation method for the overdetermined convolutive mixture model in time-domain is proposed via joint block-diagonalization based on the mutual- independence and short-time stationarity properties of the speech signals. Taking the sum of the F-norms of all off-diagonal sub-matrices as a criterion, a novel joint block-diagonalization method is proposed to estimate the whole mixture matrix through minimizing a sequence of quadratic sub-functions corresponding to mixture sub-matrices. Both theoretical analysis and simulations show that the proposed method has much lower complexity and faster convergence speed than the classical Jacobi-like method with no performance loss. In addition, there are almost no obvious impacts of the channel order and initialization values on the convergence speed.  相似文献   

2.
游荣义  陈忠 《中国物理》2005,14(11):2176-2180
Combination of the wavelet transform and independent component analysis (ICA) was employed for blind source separation (BSS) of multichannel electroencephalogram (EEG). After denoising the original signals by discrete wavelet transform, high frequency components of some noises and artifacts were removed from the original signals. The denoised signals were reconstructed again for the purpose of ICA, such that the drawback that ICA cannot distinguish noises from source signals can be overcome effectively. The practical processing results showed that this method is an effective way to BSS of multichannel EEG. The method is actually a combination of wavelet transform with adaptive neural network, so it is also useful for BBS of other complex signals.  相似文献   

3.
吕善翔  王兆山  胡志辉  冯久超 《中国物理 B》2014,23(1):10506-010506
A new method to perform blind separation of chaotic signals is articulated in this paper, which takes advantage of the underlying features in the phase space for identifying various chaotic sources. Without incorporating any prior information about the source equations, the proposed algorithm can not only separate the mixed signals in just a few iterations, but also outperforms the fast independent component analysis(FastICA) method when noise contamination is considerable.  相似文献   

4.
黄显高  Yu Pei  黄伟 《中国物理》2006,15(12):2894-2901
In this paper, a new numerical simulation approach is proposed for the study of open-loop frequency response of a chaotic masking system. Using Chua's circuit and the Lorenz system as illustrative examples, we have shown that one can employ chaos synchronization to separate the feedback network from a chaotic masking system, and then use numerical simulation to obtain the open-loop synchronization response, the phase response, and the amplitude response of a chaotic masking system. Based on the analysis of the frequency response, we have also proved that changing the amplitude of the exciting (input) signal within normal working domain does not influence the frequency response of the chaotic masking system. The new numerical simulation method developed in this paper can be extended to consider the open-loop frequency response of other systems described by differential or difference equations.  相似文献   

5.
Based on the principle of equivalent phase comparison frequency, we propose a group-period phase comparison method. This method can be used to reveal the inherent relations between periodic signals and the change laws of the phase difference. If these laws are applied in the processing of the mutual relations between frequency signals, phase comparison can be accomplished without frequency normalization. Experimental results show that the method can enhance the measurement resolution to 10-13/s in the time domain.  相似文献   

6.
A nonlinear acoustic technique for detection of contact defects in plates was developed associating guided waves with vibro-modulation techniques.The low frequency vibration varies the interface area within the contact defect;then the amplitude and phase of high frequency guided waves reflected from the contact defect will be modulated.Due to the exiting of contact defects,there was obvious low frequency component in the time-frequency distribution of modulated guided wave series.Furthermore the nonlinear responses at frequency of vibration were extracted out and used for defect imaging.It was shown that the nonlinear guided wave imaging method based on vibration modulation can effectively characterize and locate the nonlinear contact defects.  相似文献   

7.
Optical frequency domain phase conjugation(FDPC) is based on phase conjuga-tion of spectrum of an input signal.It is equivalent to the phase conjugation and the time reversal of the temporal envelope of an input signal.The use of FDPC to con-trol polarization signal distortion in birefringent optical fiber systems is proposed.Evolution of polarization signals in the system using midway FDPC is analyzed theoretically and simulated numerically.It is shown that the distortion of polariza-tion signals can be controlled effectively by FDPC.The impairments due to disper-sion and nonlinear effects can be suppressed simultaneously.  相似文献   

8.
Doppler effect widely exists in the signal from the moving acoustic source. In order to solve such problems as frequency shift and frequency band expansion, a time domain cor- rection method is presented in this paper. First, the discrete time vector for interpolation and the amplitude restoration formula is derived based on the moving relationship and the Morse acoustic theory, then the amplitude weights are corrected and the distortion signal is interpolated. Every point of the discrete signal is operated separately in time domain. Compared with the existing frequency domain methods, this method does not need to know the characteristic frequency beforehand and would not be influenced by the blending of the frequency band. Hence, this method can be employed to correct multiple frequency signals and it is also a simple and effective Doppler effect reduction method.  相似文献   

9.
A time-frequency dictionary learning approach is proposed to enhance speech contaminated by additive nonstationary noise.In this framework,a time-frequency dictionary which is learned from noise data is incorporated into the convolutive nonnegative matrix factorization framework.The update rules for the time-varying gains and speech dictionary are derived by precomputing the noise dictionary.The magnitude spectra of speech are estimated using convolution operation between the learned speech dictionary and the time-varying gains. Finally,noise is removed via binary time-frequency masking.The experimental results indicate that the proposed scheme gives better enhancement results in terms of quality measures of speech.Moreover,the proposed algorithm outperforms the multiband spectra subtraction and the non-negative sparse coding based noise reduction algorithm in nonstationary noise conditions.  相似文献   

10.
Direction of arrival estimation using array model in time-frequency domain   总被引:2,自引:1,他引:1  
Time-frequency analysis is combined with array processing to develop a direction of arrival (DOA) estimation method. The array data model is constructed in time-frequency domain by cross time-frequency distribution between the output of a reference sensor and those of two symmetric sub-arrays. Accordingly a subspace method is presented based on the average of two sub-arrays‘ time-frequency data vector model instead of the conventional array model, to estimate DOAs of multiple signals. Because the array data is processed both in spatial domain and 2-D time-frequency domain, the proposed method has an ability to select the signal of interesting, and is suitable for non-stationary signal. Additionally, the method is robust to noise and holds an advantage of low computational load. Simulations are conducted to verify the efficiency of the method and comparision is made with other methods.  相似文献   

11.
A frequency bin-wise nonlinear masking algorithm is proposed in the spectrogram domain for speech segregation in convolutive mixtures. The contributive weight from each speech source to a time-frequency unit of the mixture spectrogram is estimated by a nonlinear function based on location cues. For each sound source, a non-binary mask is formed from the estimated weights and is multiplied to the mixture spectrogram to extract the sound. Head-related transfer functions (HRTFs) are used to simulate convolutive sound mixtures perceived by listeners. Simulation results show our proposed method outperforms convolutive independent component analysis and degenerate unmixing and estimation technique methods in almost all test conditions.  相似文献   

12.
时文华  张雄伟  邹霞  孙蒙  李莉 《声学学报》2020,45(3):299-307
提出了一种联合深度编解码神经网络和时频掩蔽估计的语音增强方法。该方法利用深度编解码网络估计时频掩蔽表示,并联合带噪语音的幅度谱学习带噪语音与纯净语音幅度谱之间的非线性映射关系。深度编解码网络采用卷积-反卷积网络结构。在编码端,利用卷积网络的局部感知特性,对带噪语音的时频域结构特征进行建模,提取语音特征,同时抑制背景噪声。在解码端,利用编码端提取到的语音特征逐层恢复局部细节信息并重构语音信号。同时,在编解码端对应层之间引入跳跃连接,以减少由于池化和全连接操作导致的低层细节信息丢失的问题。在TIMIT语音库和不完全匹配噪声集下进行仿真实验,实验结果表明,该方法可以有效抑制噪声,且能较好地恢复出语音细节成分。   相似文献   

13.
Energy separation algorithm is good at tracking instantaneous changes in frequency and amplitude of modulated signals, but it is subject to the constraints of mono-component and narrow band. In most cases, time-varying modulated vibration signals of machinery consist of multiple components, and have so complicated instantaneous frequency trajectories on time-frequency plane that they overlap in frequency domain. For such signals, conventional filters fail to obtain mono-components of narrow band, and their rectangular decomposition of time-frequency plane may split instantaneous frequency trajectories thus resulting in information loss. Regarding the advantage of generalized demodulation method in decomposing multi-component signals into mono-components, an iterative generalized demodulation method is used as a preprocessing tool to separate signals into mono-components, so as to satisfy the requirements by energy separation algorithm. By this improvement, energy separation algorithm can be generalized to a broad range of signals, as long as the instantaneous frequency trajectories of signal components do not intersect on time-frequency plane. Due to the good adaptability of energy separation algorithm to instantaneous changes in signals and the mono-component decomposition nature of generalized demodulation, the derived time-frequency energy distribution has fine resolution and is free from cross term interferences. The good performance of the proposed time-frequency analysis is illustrated by analyses of a simulated signal and the on-site recorded nonstationary vibration signal of a hydroturbine rotor during a shut-down transient process, showing that it has potential to analyze time-varying modulated signals of multi-components.  相似文献   

14.
刘亚奇  刘成城  赵拥军  朱健东 《物理学报》2015,64(11):114302-114302
针对现有盲波束形成算法适用范围较窄, 多目标信号分离级联模式结构复杂、并联模式稳定性较差等问题, 提出一种基于时频分析的多目标盲波束形成算法. 该算法首先利用时频分析技术给出信号导向矢量的不确定集, 然后优化求解导向矢量的最优估计, 最后利用Capon方法实现多目标信号的并行输出. 理论分析及仿真结果表明, 该算法对信号特性没有特殊要求, 适用性较广, 性能稳定, 且输出信干噪比高于其他盲波束形成算法, 接近于最优Capon波束形成器.  相似文献   

15.
基于多项式调频Fourier变换的信号分量提取方法   总被引:1,自引:0,他引:1       下载免费PDF全文
路文龙  谢军伟  王和明  盛川 《物理学报》2016,65(8):80202-080202
为了从含有噪声的混合信号中有效提取各个信号分量, 提出一种基于多项式调频Fourier变换的分量提取方法. 通过研究Fourier变换和分数阶Fourier变换的信号能量积累方式及变换基函数的时频表示, 提出利用时频平面上的多项式调频曲线族代替Fourier变换和分数阶Fourier 变换的调频直线族, 将变换的适用范围扩展到非线性调频信号. 采用粒子群智能优化算法搜索调频曲线族的最优多项式参数, 使混合信号中的某一分量在多项式调频Fourier域上能量谱集中. 最后对能量谱集中的分量进行窄带滤波, 并利用多项式调频逆Fourier变换重构信号分量. 仿真实验结果表明, 该方法不仅能够提取混合信号中的线性调频分量, 还能够实现非线性调频分量的能量谱集中、信号分离和时频特征提取.  相似文献   

16.
A blind method for suppressing late reverberation from speech and audio signals is presented. The proposed technique operates both on the spectral and on the sub-band domains employing a single input channel. At first, a preliminary rough clean signal estimation is required and for this, any standard technique may be applied; however here the estimate is obtained through spectral subtraction. Then, an auditory masking model is employed in sub-bands to extract the reverberation masking index (RMI) which identifies signal regions with perceived alterations due to late reverberation. Utilizing a selective signal processing technique only these regions are suppressed through sub-band temporal envelope filtering based on analytical expressions. Objective and subjective measures indicate that the proposed method achieves significant late reverberation suppression for both speech and music signals over a wide range of reverberation time (RT) scenarios.  相似文献   

17.
为实现噪声情况下的人声分离,提出了一种采用稀疏非负矩阵分解与深度吸引子网络的单通道人声分离算法。首先,通过训练得到人声与噪声的字典矩阵,将其作为先验信息从带噪混合语音中分离出人声与噪声的系数矩阵;然后,根据人声系数矩阵中不同的声源成分在嵌入空间中的相似性不同,使用深度吸引子网络将其分离为各声源语音的系数矩阵;最后,使用分离得到的各语音系数矩阵与人声的字典矩阵重构干净的分离语音。在不同噪声情况下的实验结果表明,本文算法能够在抑制背景噪声的同时提高分离语音的整体质量,优于结合声噪人声分离模型的对比算法。   相似文献   

18.
徐舜  刘郁林  柏森 《应用声学》2008,27(3):173-180
盲分离算法能在缺少混合系统参数的条件下仅由观测信号估计初始源,但分离信号存在固有的排列模糊性,这往往导致两次批处理过程中同一信号"对不准",因此很难获得连续的源信号。本文针对盲声源分离中存在的相同问题,根据语音和其他音频信号的特征差异,提出一种修正的自相关函数并以其值作为一个特征基元来表征声音信号的时序相关特性,同时用平均声门波形状参数作为另一个特征基元来表征语音产生的生理效应。以这两个参数作为识别不同音频信号的二维模式特征,采用一种模糊聚类算法提取多路盲分离语音。本方法有效克服了批处理盲声源分离中的信号排列顺序的不确定性,并通过选择合适的阈值提取多路连续语音。仿真给出了5路混合音频信号中盲提取两路连续语音的实验结果。  相似文献   

19.
针对目前有监督语音增强忽略了纯净语音、噪声与带噪语音之间的幅度谱相似性对增强效果影响等问题,提出了一种联合精确比值掩蔽(ARM)与深度神经网络(DNN)的语音增强方法。该方法利用纯净语音与带噪语音、噪声与带噪语音的幅度谱归一化互相关系数,设计了一种基于时频域理想比值掩蔽的精确比值掩蔽作为目标掩蔽;然后以纯净语音和噪声幅度谱为训练目标的DNN为基线,通过该DNN的输出来估计目标掩蔽,并对基线DNN和目标掩蔽进行联合优化,增强语音由目标掩蔽从带噪语音中估计得到;此外,考虑到纯净语音与噪声的区分性信息,采用一种区分性训练函数代替均方误差(MSE)函数作为基线DNN的目标函数,以使网络输出更加准确。实验表明,区分性训练函数提升了基线DNN以及整个联合优化网络的增强效果;在匹配噪声和不匹配噪声下,相比于其它常见DNN方法,本文方法取得了更高的平均客观语音质量评估(PESQ)和短时客观可懂度(STOI),增强后的语音保留了更多语音成分,同时对噪声的抑制效果更加明显。   相似文献   

设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号