车载场景结合盲源分离与多说话人状态判决的语音抽取

在车载分布式传声器阵列场景中,结合盲源分离TRINICON (Triple-N ICA for convolutive mixtures)算法与多说话人状态判决实现期望语音抽取。根据分布式传声器阵列与声源的相对位置关系,设计特定的盲源分离初始化条件以保证输出通道与声源的映射关系;根据分布式传声器阵列的频响特点,设计特征矢量来进行多说话人判决,并将判决结果引入TRINICON算法参数迭代过程。在使用实际车载录音数据的仿真评测中,所提方法在不同信噪比下有较高的鲁棒性,可有效提升TRINICON算法的收敛速度和语音信号的信扰比,且可以确保准确的通道映射。评测结果表明该方法可以在车载场景中有效抽取出期望语音,为车载复杂场景下的声信息提取提供了一种可靠且收敛快速的解决方法。      

基于声源方位信息和非线性时频掩蔽的语音盲提取算法

针对欠定卷积混合的语音信号模型,提出一种基于声源方位信息和非线性时频掩蔽的语音盲提取算法。首先对低频段混合语音信号进行时频分析估计瞬时相对时延(ITD)并采用势函数聚类分析方法估计出声源个数及其ITD,接着锁定目标提取准确的目标语音方位信息,最后利用独立语音在时频域上的近似W一分离正交性,采用非线性时频掩蔽的方法提取目标语音。仿真实验表明,该方法能锁定任意感兴趣目标方位,能有效提取目标语音,文中实验条件下信噪比增益平均达9.5 dB。     

扩散噪声下协方差矩阵重构的语音分离与降噪

针对传统多通道语音分离算法在扩散噪声下性能下降的问题,提出了一种用于语音分离及降噪的空间协方差模型及参数估计方法。该方法将扩散噪声视为独立声源,利用由导向矢量重构的空间协方差矩阵建模目标声源的空间特性,并通过空间协方差分析方法估计用于语音分离的多通道维纳滤波器。同时,还提出了一种联合该方法的后置滤波器参数框架,为输出信号降噪和失真的折中提供了更多选择。在扩散噪声下的单目标和多目标实验中,所提方法的语音提取和分离性能都优于对比算法,联合参数的后置滤波器可提供更为符合人们要求的降噪语音,验证了所提模型与参数估计方法的有效性。      

基于总体最小二乘算法的平稳声信号二阶盲分离方法

利用声源信号相互独立的已知条件,提出分离平稳声信号的总体最小二乘二阶盲分离算法,该算法在联合近似对角化相关函数矩阵的约束下估计分离矩阵,在同时考虑估计矩阵与观测信号误差的情况下给出修正模型,在修正模型总体误差最小时分离源信号。提出的算法具有无需选择非线性函数,不存在收敛到局部最小值与计算速度快的特点,分离的声源信号具有较小的失真。半消音室电机噪声分离实验验证了上述算法的有效性。     

基于平板冲击声的声源特性表征及自动识别*

以平板结构导纳函数为纽带,建立冲击声信号特征与声源特性之间的关联,获得与声源属性密切相关的特征用于目标分类。针对四边简支矩形被击板,借助信号参数识别算法获得与声源物理属性有关的6维导纳特征,并从冲击声样本中提取80维音色特征,将音色特征和导纳特征做相关性分析,获得与声源物理属性相关的信号特征集。利用BP神经网络进行分类,结果表明,当采用与特定声源物理属性相关的信号特征子集时,分类效果达到同组最优。     

子带t分布的快速独立向量分析在语声盲源 分离中的应用*

为了提高独立向量分析算法在盲语声分离任务中的分离性能,降低算法计算复杂度,并改善目前尚未完全解决的顺序模糊性的问题,该文提出一种基于子带t分布的快速独立向量分析算法。在声源模型方面,该算法首先利用语声信号重尾分布的特性,假设声源概率密度函数服从t分布,同时采用子带建模的方法来增强同一声源相邻频点的高阶依赖性,进而减轻频点间的顺序不一致问题。在空间模型方面,该算法采用秩1更新的方式估计声源信号,避免矩阵求逆操作和分离矩阵的估计,从而降低计算复杂度。实验结果表明,与现有的基于独立向量分析的盲源分离算法相比,该算法能够在相同的迭代次数下取得更优的语声分离性能。     

一种适用于微弱信号提取的盲源分离算法

本文首次提出了适用于微弱信号提取的盲源分离算法,这种方法是在常用的自然梯度串行更新算法基础上采用了有监督的机制。通过观察两个信号互相关的程度自动判别激活函数的种类,实现了超高斯亚高斯强弱信号混合下的盲源分离。信号仿真表明,有监督的盲源分离技术收敛速度较快,精度较高。     

基于最大信噪比测量两相流速度

为了克服噪声对信号的影响,提出一种利用最大信噪比和相关法测量两相流速度的方法.基于最大信噪比的信号分离方法是一种盲源信号分离方法,该算法利用统计独立信号完全分离时信噪比最大作为分离准则,它具有非常低的计算复杂度.这里首先利用盲源信号分离方法分别提取出上游和下游两相流信号,并据此求出两相流信号的相关函数曲线,由此求出信号的渡越时间,最后给出仿真实验的处理结果.实验结果表明该方法能够满足两相流速度的测量要求.     

一种利用分布式传声器阵列的声源三维定位方法

为了提高噪声和混响条件下分布式传声器阵列进行声源定位的性能,提出一种利用空间稀疏性和压缩感知原理的声源三维定位方法。该方法首先通过两次离散余弦变换方式提取出声音信号特征,并用该特征来构建稀疏定位模型,以便能够综合利用语音信号的短时和长时特性,同时降低模型维数;然后利用在线字典学习技术动态调整字典,克服稀疏模型与实际信号之间的失配问题,增强稀疏定位模型的鲁棒性;进而提出一种改进的平滑l₀范数稀疏重构算法来进行声源位置解算,以提高低信噪比条件下的重构精度。仿真结果表明该方法不仅可以实现多目标定位,而且具有较强的抗噪声和抗混响能力.      

交替投影盲反卷积的阵不变量被动定位*

本文针对低信噪比下传统盲反卷积技术提取格林函数过程中存在的图像模糊问题,提出了一种基于交替投影法的阵不变量浅海被动定位方法。该种方法利用交替投影的方法对波束信号进行分离,并利用分离出的波束分别提取格林函数。最后通过补偿时延的方式将多组波束提取的结果相干组合起来,实现格林函数的多波束联合估计。通过仿真实验证明了该方法有效增强了提取格林函数的强度,使得在低信噪比下利用阵不变量的连续声源定位方法更加稳健。     

Blind speech source separation via nonlinear time-frequency masking

Aim at the underdetermined convolutive mixture model, a blind speech source separation method based on nonlinear time-frequency masking was proposed, where the approximate W-disjoint orthogonality （W-DO） property among independent speech signals in time-frequency domain is utilized. In this method, the observation mixture signal from multimicrophones is normalized to be independent of frequency in the time-frequency domain at first, then the dynamic clustering algorithm is adopted to obtain the active source information in each time-frequency slot, a nonlinear function via deflection angle from the cluster center is selected for time-frequency masking, finally the blind separation of mixture speech signals can be achieved by inverse STFT （short-time Fourier transformation）. This method can not only solve the problem of frequency permutation which may be met in most classic frequency-domain blind separation techniques, but also suppress the spatial direction diffusion of the separation matrix. The simulation results demonstrate that the proposed separation method is better than the typical BLUES method, the signal-noise-ratio gain （SNRG） increases 1.58 dB averagely.     

噪声情况下采用稀疏非负矩阵分解与深度吸引子网络的人声分离算法

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。      

一种基于人工蜂群算法的混沌信号盲分离方法

混沌信号所固有的非周期、宽带频谱和对初值极度敏感等特性使得对这类信号进行盲分离极为困难. 针对这一问题, 提出一种新的盲分离方法, 该方法通过相空间重构来构造代价函数, 将混沌信号的盲分离转化为一个无约束优化问题, 并利用人工蜂群算法进行求解. 不同于现有的独立成分分析方法仅使用混合信号的统计特性来解决分离问题, 该方法能充分利用混合信号内在的动态特性, 因而在处理混沌信号这种确定性信号时能获得更好的分离效果. 此外, 正交矩阵的参数化表示有效地降低了盲分离问题的复杂性, 使优化过程能快速收敛. 实验结果表明, 该方法具有较快的收敛速度和较高的数值精度, 在分离混沌信号时其整体性能优于现有的几种盲分离方法. 同时, 在分离混沌-高斯混合信号的实验中该方法也展现出优异良好的性能, 这表明该文的方法有应用潜力.     

Blind convolutive separation method for speech signals via joint block diagonalization

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.     

Blind Extraction of Chaotic Signals by Using the Fast Independent Component Analysis Algorithm

We report the results of using the fast independent component analysis （FastICA） algorithm to realize Mind extraction of chaotic signals. Two cases are taken into consideration： namely, the mixture is noiseless or contaminated by noise. Pre-whitening is employed to reduce the effect of noise before using the FastICA algorithm. The correlation coefficient criterion is adopted to evaluate the performance, and the success rate is defined as a new criterion to indicate the performance with respect to noise or different mixing matrices. Simulation results show that the FastICA algorithm can extract the chaotic signals effectively. The impact of noise, the length of a signal frame, the number of sources and the number of observed mixtures on the performance is investigated in detail It is also shown that regarding a noise as an independent source is not always correct.     

Time-delay estimation of acoustic emission signals using ICA

Acoustic emission (AE) analysis is used for characterization and location of developing defects in materials. AE sources often generate a mixture of various statistically independent signals. One difficult problem of AE analysis is the separation and characterization of signal components when the signals from various sources and the way in which the signals were mixed are unknown. Recently, blind source separation by independent component analysis (ICA) has been used to solve these problems. The main purpose of this paper is to demonstrate the applicability of ICA to time-delay (T-D) estimation of two independent continuous AE sources on an aluminum beam. It is shown that it is possible to estimate T-Ds by ICA, and thus to locate two independent simultaneously emitted sources.     

非线性混合模式的语音盲分离算法

本文针对FIR非线性混合模型，基于最大熵算法，提出了一种以高斯混合模式概率密度函数估计替代传统对数化概率密度估计的盲分离算法，以偶函数为非线性激活函数，采用最大期望（EM）迭代算法推导了分离算法的权向量迭代公式，通过模拟仿真实验结果与传统的最大熵和高阶累积量方法比较，新算法提高了收敛速度，并有效地完成了非线性语音分离任务，抑制了干扰语音信号的影响，提高了输出信噪比。     

A new regularized forward blind source separation algorithm for automatic speech quality enhancement

This paper addresses the problem of speech enhancement and acoustic noise reduction by adaptive filtering algorithms in a moving car through blind source separation (BSS) structures. In this paper we propose a new regularized forward blind source separation (RFBSS) algorithm that does not need voice activity detection (VAD) systems, and allows getting efficient speech enhancement performances with low complexity.     

A blind separation algorithm for underdetermined convolutional mixed communication signals based on time–frequency soft mask

This paper focuses on the separation for time–frequency (TF) overlapped communication signals received by the sensors. A novel blind separation strategy is proposed to improve the poor performance of signal separation by traditional algorithms for convolutional mixtures in underdetermined cases. Firstly, the number of sources and cluster centers are obtained in the sparse domain by combining the density peak clustering (DPC) with fuzzy c-means (FCM) clustering algorithm; Then the GMM clustering algorithm is applied to calculate the membership degree of the source signal in the mixed signals, so as to construct a TF soft mask matrix to more precisely carry out separation for TF overlapped signals. In this paper, the separation simulations are conducted with the digital modulation signals of 2ASK, BPSK, QPSK, etc. The results show that the algorithm proposed in this paper has better anti-aliasing and anti-noise performance than the comparison algorithms.     

分布式压缩感知麦克风阵列多声源方位估计*

麦克风阵列已被广泛应用于音/视频会议等人机交互领域中时，多声源应用场景对声源方位估计性能提出了更高的要求。压缩感知(CS)声源定位算法将声源定位问题转化为信号的稀疏重构问题，相比传统的定位算法如相位变换加权（SRP-PHAT）和时延累加定位（DS）能够获得较高的定位性能，但多声源的存在一定程度上降低了稀疏程度，影响了CS重构性能。考虑到传统的CS定位算法并未利用多个连续语音帧之间声源空间向量的共同稀疏性，提出采用分布式压缩感知（DCS）理论以改善多声源的稀疏恢复估计的性能。仿真和实验结果表明，相比于传统定位算法和CS-OMP算法，DCS-SOMP算法在不同信噪比和不同声源强度的环境中，对多声源的方位估计都具有更好的定位性能和定位稳健性。