A wavelet-based forward BSS algorithm for acoustic noise reduction and speech enhancement

In this paper, we address the problem of noise reduction and speech enhancement by adaptive filtering algorithm. Recently, the well known forward blind source separation (FBSS) structure has been largely studied and intensively used to reduce acoustic noise components and to enhance speech signal. The FBSS structure is often combined with adaptive algorithms to accelerate the adaptation of the cross-filters, and to improve noise suppression at the output. In this paper, we propose to use a wavelet transform decomposition in the FBSS structure by using a two-channel forward wavelet symmetric adaptive decorrelating (WFSAD) algorithm. The proposed WFSAD algorithm provides a better compromise between time and frequency resolution and improves robustness of the noise reduction process when compared with the classical two-channel forward symmetric adaptive decorrelating (FSAD) algorithm. Simulation results prove the efficiency of the proposed WFBSS algorithm in comparison with conventional ones in terms of several objective and subjective criteria.     

小波包自适应阈值语音降噪新算法

为了克服低信噪比输入下,语音增强造成语音清音中的弱分量损失,造成重构信号包络失真的问题。论文提出了一种新的语音增强方法。该方法根据语音感知模型,采用不完全小波包分解拟合语音临界频带,并对语音按子带能量进行清浊音区分处理,在阈值计算上,提出了一种清浊音分离,基于子带信号能量的小波包自适应阈值算法。通过仿真实验,客观评测和听音测试表明,该算法在低信噪比输入时较传统算法,能够更加有效地减少重构信号包络失真,在不损伤语音清晰度和自然度的前提下,使输出信噪比明显提高。将该算法与能量谱减法结合,进行二次增强能进一步提高降噪输出的语音质量。     

听觉频域掩蔽效应的自适应β阶贝叶斯感知估计语音增强算法

为了在噪声抑制和语音失真中之间寻找最佳平衡,提出了一种听觉频域掩蔽效应的自适应β阶贝叶斯感知估计语音增强算法,以期提高语音增强的综合性能。算法利用了人耳的听觉掩蔽效应,根据计算得到的频域掩蔽阈自适应调整β阶贝叶斯感知估计语音增强算法中的β值,从而仅将噪声抑制在掩蔽阈之下,保留较多的语音信息,降低语音失真。并分别用客观和主观评价方式,对所提出的算法的性能进行了评估,并与原来基于信噪比的自适应β阶贝叶斯感知估计语音增强算法进行了比较。结果表明,频域掩蔽的β阶贝叶斯感知估计方法的综合客观评价结果在信噪比为-10 dB至5 dB之间时均高于基于信噪比的自适应β阶贝叶斯感知估计语音增强算法。主观评价结果也表明频域掩蔽的β阶贝叶斯感知估计方法能在尽量保留语音信息的同时,较好的抑制背景噪声。      

语音通信降噪研究

语音通信系统中，语音通过信道传输将不可避免地引入码间串扰和信号畸变，同时受到噪声污染。本文在分析自适应盲均衡算法CMA(constant modulus algorithm）和改进盲均衡算法的基础上，考虑到自适应盲均衡技术在语音噪声控制方面能力有限，将自适应盲均衡技术与小波包掩蔽阈值降噪算法联合使用，形成一种基带语音增强新方法。仿真试验结果显示自适应盲均衡技术可以使星座图变得清晰而紧凑，有效减小误码率。研究证实该方法在语音信号ISI和畸变严重情况下，在白噪及有色噪声不同的噪声环境中都具有稳定的降噪能力，消噪同时可获得汉语普通话良好的听觉效果。     

Second generation wavelet transform-based pitch period estimation and voiced/unvoiced decision for speech signals

Pitch detection is an important part of speech recognition and speech processing. In this paper, a pitch detection algorithm based on second generation wavelet transform was developed. The proposed algorithm reduces the computational load of those algorithms that were based on classical wavelet transform. The proposed pitch detection algorithm was tested for both real speech and synthetic speech signal. Some experiments were carried out under noisy environment condition to evaluate the accuracy and robustness of the proposed algorithm. Results showed that the proposed algorithm was robust to noise and provided accurate estimates of the pitch period for both low-pitched and high-pitched speakers. Moreover, different wavelet filters that were obtained using second generation wavelet transform were considered to see the effects of them on the proposed algorithm. It was noticed that Haar filter showed good performance as compared to the other wavelet filters.     

结合幅度谱和功率谱字典的语音增强方法

从双路字典学习、噪声功率谱估计、语音幅度谱重构角度提出了一种改进的谱特征稀疏表示语音增强方法。在字典学习阶段,融合功率谱与幅度谱特征,采用区分性字典降低语音字典和噪声字典的相干性;在语音增强阶段,提出一种噪声功率谱估计方法对非平稳噪声进行跟踪估计;考虑到幅度谱和功率谱特征对不同噪声的适应程度不同,设计了语音重构权值表。对分别由幅度谱和功率谱恢复而来的两路信号进行自适应加权重构,结合相位补偿函数得到增强后的语音信号。实验结果表明,该方法在平稳、非平稳噪声环境下相比于单一谱特征的语音增强方法平均提高31.6%,改善了语音增强方法的性能。      

Speech enhancement method based on low-rank approximation in a reproducing kernel Hilbert space

Speech signal is corrupted unavoidably by noisy environment in subway, factory, and restaurant or speech from other speakers in speech communication. Speech enhancement methods have been widely studied to minimize noise influence in different linear transform domain, such as discrete Fourier transform domain, Karhunen-Loeve transform domain or discrete cosine transform domain. Kernel method as a nonlinear transform has received a lot of interest recently and is commonly used in many applications including audio signal processing. However this kind of method typically suffers from the computational complexity. In this paper, we propose a speech enhancement algorithm using low-rank approximation in a reproducing kernel Hilbert space to reduce storage space and running time with very little performance loss in the enhanced speech. We also analyze the root mean squared error bound between the enhanced vectors obtained by the approximation kernel matrix and the full kernel matrix. Simulations show that the proposed method can improve the computation speed of the algorithm with the approximate performance compared with that of the full kernel matrix.     

Speech enhancement based on the discrete Gabor transform and multi-notch adaptive digital filters

This paper presents a new method to speech enhancement based on time-frequency analysis and adaptive digital filtering. The proposed method for dual-channel speech enhancement was developed by tracking frequencies of corrupting signal by the discrete Gabor transform (DGT) and implementing multi-notch adaptive digital filter (MNADF) at those frequencies. Since no a priori knowledge of the noise source statistics is required this method differs from traditional speech enhancement methods. Specifically, the proposed method was applied to the case where speech quality and intelligibility deteriorate in the presence of background noise. Speech coders and automatic speech recognition (ASR) systems are designed to act on clean speech signals. Therefore, corrupted speech signals by the noise must be enhanced before their processing. The method uses a primary input containing the corrupted speech signal while a reference input containing the noise only. In this paper, we designed MNADF instead of single-notch adaptive digital filter and used DGT to track frequencies of corrupting signal because fast filtering process and fast measure of the time-dependent noise frequency are of great importance in speech enhancement process. Therefore, MNADF was implemented to take advantage of fast filtering process. Different types of noises from Noisex-92 database were used to degrade real speech signals. Objective measures, the study of the speech spectrograms and global signal-to-noise ratio (SNR), segmental SNR (segSNR), Itakura-Saito distance measure as well as subjective listing test demonstrated consistently superior enhancement performance of the proposed method over traditional speech enhancement method such as spectral subtraction. Combining MNADF and DGT, excellent speech enhancement was obtained.     

Narrowband speech wideband extension algorithm research

To reduce the spectral distortion,a Hidden Markov Model-based narrowband speech bandwidth extension algorithm is presented.Firstly,the parameters which have higher mutual information with wideband envelope were extracted to constitute the feature vector,and then a posterior probability was calculated via the joint probability of the partial observation feature vector sequence and the markov states.Secondly,based on the posterior probability,the wideband envelope was estimated using Bayesian parameter estimation method and minimum mean square error criteria.For estimation of wideband excitation signal,intermediate frequency extension algorithm is proposed based on the harmonic correlation between the low frequency and high frequency.The experimental results show that,compared with the traditional bandwidth extension algorithm based on Hidden Markov Model,the average spectral distortion is reduced by 0.187 dB and the number of speech frame with spectral distortion over10dB is decreased by 34.3%.     

Noise estimation based on time–frequency correlation for speech enhancement

As a fundamental part of speech enhancement, noise estimation is particularly challenging in highly non-stationary noise environments. In this work, we propose an effective algorithm on the basis of the “Improved Minima Controlled Recursive Averaging (IMCRA)” with the objective to improve the performance of noise estimation. The main contributions of this work are: (i) in the algorithm, a rough decision about speech presence is proposed by calculating the autocorrelation and cross-channel correlation of the T–F (Time–Frequency) units; (ii) with this decision, we refine the smoothing parameters for the smoothing of noisy power spectrum and the recursive averaging in noise spectrum estimation as well as the weighting factor for the a priori SNR (Signal to Noise Ratio) estimation in the IMCRA; (iii) we improve the search of local minima during spectral bursts by adding a minimum search with a shorter window. Extensive experiments are carried out to evaluate the performance of our proposed algorithm. The experimental results illustrate that, compared with the IMCRA, the proposed approach significantly improves the accuracy of noise spectrum estimation and the quality of enhanced speech in the typical noise situations.     

Global and local contrast enhancement algorithm for image using wavelet neural network and stationary wavelet transform

A new contrast enhancement algorithm for image is proposed employing wavelet neural network (WNN)and stationary wavelet transform (SWT). Incomplete Beta transform (IBT) is used to enhance the global contrast for image. In order to avoid the expensive time for traditional contrast enhancement algorithms,which search optimal gray transform parameters in the whole gray transform parameter space, a new criterion is proposed with gray level histogram. Contrast type for original image is determined employing the new criterion. Gray transform parameter space is given respectively according to different contrast types,which shrinks the parameter space greatly. Nonlinear transform parameters are searched by simulated annealing algorithm (SA) so as to obtain optimal gray transform parameters. Thus the searching direction and selection of initial values of simulated annealing is guided by the new parameter space. In order to calculate IBT in the whole image, a kind of WNN is proposed to approximate the IBT. Having enhanced the global contrast to input image, discrete SWT is done to the image which has been processed by previous global enhancement method, local contrast enhancement is implemented by a kind of nonlinear operator in the high frequency sub-band images of each decomposition level respectively. Experimental results show that the new algorithm is able to adaptively enhance the global contrast for the original image while it also extrudes the detail of the targets in the original image well. The computation complexity for the new algorithm is O(MN) log(MN), where M and N are width and height of the original image, respectively.     

基于二次相关的语音信号时延估计改进算法

目前语音信号的时延估计研究,大部分采用的是广义互相关算法。然而,广义互相关时延估计算法易受噪声和混响环境影响。为此,本文提出了一种基于二次相关的语音信号时延估计改进算法,该算法对语音信号进行二次互相关运算,并结合Hilbert变换,对二次互相关峰值进行进一步的锐化处理,使得反映时延的峰值点检测更为准确。实验结果表明,改进的时延估计方法在非平稳的语音信号中能够有效地抑制噪声干扰,且在不同混响条件下时延估计具有更好的性能。     

基于小波模极大值移位相关的光谱去噪方法

提出一种小波模极大值移位相关滤波算法,用来对吸收光谱进行去噪处理。首先,依据小波变换模极大值理论,识别吸收光谱的二进小波系数中的噪声成分和有用信号成分;然后将其中相邻尺度的有用成分在波数轴上移位对齐,修正模极大值在尺度间的“漂移”现象,并对其中噪声成分进行平滑;最后将预处理后的相邻尺度的小波子带相乘,根据得到的小波系数尺度间的相关度,进一步强化信号的重要特征,同时衰减噪声。相比SSNF及MPTH等去噪方法,新的算法不需要估计噪声强度,可避免引入误差和复杂计算;也不需要进行迭代计算,可消除算法收敛缓慢甚至不收敛的危险;同时,该算法修正了光谱谱峰在不同尺度的模极大值位置的“漂移”,可弥补这一现象造成的谱信息的损失。实验结果证明,提出的滤波算法在保留SF₆气体红外吸收光谱的有用成分的前提下,有效的滤除了噪声。     

小波相位分析测量成像径向畸变

为了测量光学成像像面各个像素的径向畸变大小,提出将小波变换载频条纹相位分析应用于径向畸变测量。采用正弦载波条纹作为测量模板,把径向畸变转化为径向调制相位。应用条纹相位分析导出径向调制相位和径向畸变的转化关系。采用小波频率估计和相位估计提取变形条纹的相位,由于变形条纹中心点是零畸变,中心点的瞬时频率和相位可以计算参考条纹的基频相位。两种基频相位之差就是与所有像素径向位置畸变分布对应的三维调制相位——称为径向畸变分布。利用校正公式和立方卷积插值算法对彩色畸变图像进行校正,给出详细的理论分析和实验结果。     

基于小波变换与卷积神经网络的图像去噪算法

图像在生成或传感过程中往往会受到噪声干扰，噪声干扰会给后续图像处理工作增加难度，甚至会给某些生产活动带来巨大的经济损失。结合平稳小波变换与卷积神经网络的优势，提出了一种有效的图像去噪算法。训练阶段，采用提出的算法对图像进行尺度为1的平稳小波分解后，分别把高、低频分量输入4个设计好的残差网络进行训练；在测试阶段使用小波逆变换来获得最终的预测图像。实验结果表明:在高斯白噪声水平达到σ=50时，去噪后图像的峰值信噪比（peak signal to noise ratio, PSNR）均值和结构相似性（structural similarity index method, SSIM）均值可以达到28.37 dB和0.808 0，提出的算法可以有效去除可见光图像中的高斯白噪声、自然噪声，以及遥感图像在传感过程中产生的噪声，并且在去除图像噪声的同时能较好地保留图像的边缘与纹理细节。     

基于听觉掩蔽效应的 阶STSA-MMSE语音增强算法

提出人耳掩蔽效应与阶STSA- MMSE（Short Time Spectral Amplitude-Minimum Mean Square Error）算法动态结合的语音增强算法.该算法通过引入参量提高了STSA-MMSE 算法的实时性,同时结合人耳掩蔽效应,动态的确定增强滤波器的传递函数以适应语音信号的变化,来提高语音质量.实验结果表明,和STSA-MMSE 算法相比,该算法在实时性方面有很大改善,并使降噪后的语音信号有较小的失真,同时很好地抑制了音乐噪音.     

融合Retinex和离散小波奇异值分解的远距离目标图像清晰化

针对远距离成像系统获取的低照度降质图像增强问题,提出了一种融合Retinex和离散小波奇异值分解的图像清晰化算法。该方法首先利用自适应全尺度Retinex（adaptive full-scale retinex, AFSR）“粗”提取照度分量和反射分量,然后通过离散小波变换将所提取的图像反射分量分解为4个频率子带并估计出低频子带图像的奇异值矩阵,最后应用逆小波变换“精”重建图像。实验结果表明:所提方法处理后的低照度降质图像视觉增强效果较好,在图像对比度、信息熵、平均梯度和边缘密度等客观评价指标方面优于其他经典算法。     

基于中值滤波和提升小波分析的图像去噪方法研究

针对现有算法大多对单一高斯噪声或脉冲噪声进行图像滤波的问题,在对二维图像平滑去噪的过程中,采用基于中值滤波和提升小波变换相结合的图像去噪方法。在中值滤波基础上,构造基于脉冲检测的中值滤波器,找出混合噪声中脉冲噪声并进行滤波;与此同时,对原始小波进行提升,构造提升小波,然后采用提升小波阈值去噪方法抑制高斯噪声。实验结果表明:采用本文方法,混合噪声得到有效抑制,去噪效果好。     

窄带语音带宽扩展算法研究

为了降低谱失真,提出了一种基于隐马尔科夫模型的窄带语音带宽扩展算法。首先,算法选取与宽带谱包络互信息大的参数构成特征矢量,并利用隐马尔可夫状态和过去观察特征矢量的联合先验概率估计条件后验概率。其次,以条件后验概率为基础,算法结合贝叶斯条件参数估计法和最小均方差准则估计宽带谱包络。针对宽带激励信号估计,基于信号高频和低频的谐波相关性,提出了一种中频激励扩展算法。实验结果表明,与传统的基于隐马尔可夫模型的带宽扩展算法相比,本文算法可降低0.187 dB的平均谱失真,将谱失真大于10 dB的语音帧减少了34.3%。      

基于提升小波变换的阈值改进去噪算法在紫外可见光谱中的研究

在紫外可见光谱定量分析中,由于分光光度计内部的光学系统、光源、检测器、电子元器件,电路设计以及外部环境干扰等因素产生的随机噪声,严重影响光谱定量分析结果的准确性,为提高紫外可见光谱分析精度,需要对光谱数据进行去噪预处理。由于小波分析具有多分辨率,低熵性、去相关性等特点,基于小波分析的去噪算法优于传统的去噪算法,目前基于小波去噪的方法主要有模极大值去噪算法,系数相关去噪算法,阈值去噪算法,工程实际应用以Donoho的阈值去噪法最为常用。根据Donoho阈值消噪原理,提出一种基于提升小波变换的阈值改进算法,一方面使用提升小波变换,提升小波变换是第二代小波变换,继承了小波的多分辨率特性,并且不需要进行傅里叶变换,从而具有算法简单,速度快,实现简单的优点;另一方面提出了一种新的阈值函数,克服了硬阈值函数在阈值处不连续以及软阈值函数存在恒定偏差的问题,同时对阈值估计进行了调整,有利于信号小波系数的保留和噪声小波系数的剔除。对三组多金属离子混合溶液的实测紫外可见光谱信号,添加随机噪声后使用该方法进行去噪处理,并使用信噪比(SNR)和均方根误差(RMSE)进行去噪性能评价。试验结果表明,提出的算法优于Donoho的软硬阈值去噪算法,能够有效提高光谱信噪比和降低均方根误差,从而更好地消除光谱信号中的噪声和保留光谱信号中一些重要的细节特征,比较适合用于紫外可见光谱数据建模之前的去噪预处理,在紫外可见光谱信号分析中具有较好的应用前景。