多通道助听器语音降噪算法研究

维纳滤波算法是改善噪声环境下听障患者语音理解度的常用算法之一。针对传统维纳滤波算法噪声谱估计偏差大的问题,提出一种基于改进的多通道维纳滤波算法的助听器语音降噪算法。算法首先结合人耳听觉特性和助听器响度补偿的特点,将语音信号进行Gammatone分解为多路子带信号。然后在每个子带内用基于先验信噪比估计的维纳滤波器进行语音增强处理。最后通过综合子带信号,得到增强的语音。此外,为了改善维纳滤波算法噪声谱估计的问题,提出一种基于包络估计的语音活动检测算法,并用于改善维纳滤波性能。实验结果表明,与传统维纳滤波法相比,该方法能更有效地抑制残留噪声,提高语音可懂度,具有较高的实用价值。     

基于小波域维纳滤波的图像降噪技术

传统的小波域局部维纳滤波器的参数由小波系数的某个邻域上的系数或某个邻域上的系数加上相邻尺度上的对应系数所估计,由于邻域不可能取得很大,这样会导致在某些点上估计精度的下降.对此,该文首先分析了传统的小波域局部维纳滤波器的估计误差,然后根据分析的结果,提出了一种对该算法的改进,即先用适当的门限值对小波系数进行阈值化处理,再进行局部维纳滤波.对不同噪声水平的测试图像的仿真结果表明,该改进措施可以有效地改善小波域局部维纳滤波的降噪性能,而且噪声污染越严重,改善越明显.     

一种图像自适应Wiener组合滤波方法

利用小波域Wiener滤波和空间域自适应Wiener滤波的特点,提出一种基于小波域自适应Wiener滤波和空间域自适应Wiener滤波的组合滤波方法。该方法首先在小波域进行自适应Wiener滤波,对恢复图像中的残留噪声方差进行重新估计,再在空间域进行自适应Wiener滤波,这种方法提高了恢复图像的精度。仿真实验表明,与单独的小波域和空间域Wiener滤波相比,该方法的均方误差最小,去噪效果更优。     

基于卷积神经网络的面罩语音识别

针对带噪面罩语音识别率低的问题,结合语音增强算法,对面罩语音进行噪声抑制处理,提高信噪比,在语音增强中提出了一种改进的维纳滤波法,通过谱熵法检测有话帧和无话帧来更新噪声功率谱,同时引入参数控制增益函数;提取面罩语音信号的Mel频率倒谱系数(MFCC)作为特征参数;通过卷积神经网络(CNN)进行训练和识别,并在每个池化层后经局部响应归一化(LRN)进行优化.实验结果表明:该识别系统能够在很大程度上提高带噪面罩语音的识别率.     

Image denoising algorithm via best wavelet packet base using Wiener cost function

A new cost function, namely, the Wiener cost function, is introduced to find the best wavelet packet (WP) base in image denoising. Unlike the existing entropy-type cost functions in image compression, the Wiener cost function depends on both sparseness of image representation and noise level. Combining the Wiener cost function and the doubly local Wiener filtering scheme, a new image denoising algorithm is proposed using the best wavelet packet bases. Owing to unknown true image in denoising, a pilot image with less noise is required to find the best wavelet packet base, which is obtained by the existing denoising algorithms. From the pilot image, the best 2D wavelet packet tree is searched in terms of the Wiener cost function and the energy distributions of the image in the best wavelet packet domain are also estimated. Further, the image is recovered by applying the local Wiener filtering to the best wavelet packet coefficients of the noisy image. The experimental results show that for images of structural textures, for example 'Barbara' and texture images, the proposed algorithm greatly improves denoising performance as compared with the existing state-of-the-art algorithms.     

DCT域维纳滤波语音增强

针对非平稳噪声和强背景噪声下声音信号难以提取的实际问题,提出了一种DCT域的维纳滤波方法。列出了DCT域清浊音分割步骤,给出了DCT域频谱信噪比迭代更新机制与具体实施方案,设计了DCT域的二维维纳滤波。实验仿真表明,该算法能有效地去噪滤波,改善可懂度,且在不同的噪声环境和信噪比条件下具有鲁棒性。该算法计算代价小,简单易实现。     

Car Speech Enhancement Using a Microphone Array

This paper proposes a speech enhancement approach to suppress the interference of car noise. A linear microphone array is adopted for far-talking speech acquisition and delay-and-sum beamforming noise reduction. We present an effective time delay estimator using the coherence function between the reference microphone and the beamformed speech. To further enhance the beamformed speech, we exploit an improved Wiener filter where the resulting noise correlation in microphone array is relatively small so that the performance of optimal Wiener filtering could be achieved. Also, due to the serious degradation in low frequency car speech, we develop a spectral weighting function to compensate the low frequency filtering. These two processing units serve as the post filters to attain the desirable enhancement performance. In the experiments on microphone array speech in presence of real and simulated car noises, we find that the proposed algorithm performs well. Performance is measured in terms of the signal-to-noise ratio and the word error rate. The combined delay-and-sum beamformer and two post filters obtain the best results compared to other methods.     

Speech enhancement based on stationary bionic wavelet transform and maximum a posterior estimator of magnitude-squared spectrum

Numerous efforts have focused on the problem of reducing the impact of noise on the performance of various speech systems such as speech coding, speech recognition and speaker recognition. These approaches consider alternative speech features, improved speech modeling, or alternative training for acoustic speech models. In this paper, we propose a new speech enhancement technique, which integrates a new proposed wavelet transform which we call stationary bionic wavelet transform (SBWT) and the maximum a posterior estimator of magnitude-squared spectrum (MSS-MAP). The SBWT is introduced in order to solve the problem of the perfect reconstruction associated with the bionic wavelet transform. The MSS-MAP estimation was used for estimation of speech in the SBWT domain. The experiments were conducted for various noise types and different speech signals. The results of the proposed technique were compared with those of other popular methods such as Wiener filtering and MSS-MAP estimation in frequency domain. To test the performance of the proposed speech enhancement system, four objective quality measurement tests [signal to noise ratio (SNR), segmental SNR, Itakura–Saito distance and perceptual evaluation of speech quality] were conducted for various noise types and SNRs. Experimental results and objective quality measurement test results proved the performance of the proposed speech enhancement technique. It provided sufficient noise reduction and good intelligibility and perceptual quality, without causing considerable signal distortion and musical background noise.     

基于Contourlet变换和Wiener滤波的图像降噪

提出一种新的基于Contourlet变换和Wiener滤波的图像降噪方法。该方法充分利用Contourlet变换域系数服从广义高斯分布的特点,在Contourlet域采用Bayes收缩阈值法进行预降噪,采用Wiener滤波法对预降噪图像中的残留噪声进行进一步处理,以提高图像的恢复精度。仿真结果表明,该方法较传统的Contourlet域降噪方法具有更好的降噪效果,进一步提高了PSNR值,降低了MSE值,获得了更好的图像恢复质量。     

Single channel noise reduction system in low SNR

We propose a two stage noise reduction system for reducing background noise using single-microphone recordings in very low signal-to-noise ratio (SNR) based on Wiener filtering and ideal binary masking. The proposed system contains two stages. In first stage, the Wiener filtering with improved a priori SNR is applied to noisy speech for background noise reduction. In second stage, the ideal binary mask is estimated at every time–frequency channel by using pre-processed first stage speech and comparing the time–frequency channels against a pre-selected threshold T to reduce the residual noise. The time–frequency channels satisfying the threshold are preserved whereas all other time–frequency channels are attenuated. The results revealed substantial improvements in speech intelligibility and quality over that accomplished with the traditional noise reduction algorithms and unprocessed speech.     

含噪语音实时迭代维纳滤波

针对传统去噪方法在强背景噪声情况下,提取声音信号的能力变弱甚至失效与对不同噪声环境适应性差,提出了迭代维纳滤波声音信号特征提取方法。给出了语音噪声频谱与功率谱信噪比迭代更新机制与具体实施方案。实验仿真表明,该算法能有效地去噪滤波,显著地提高语音识别系统性能,且在不同的噪声环境和信噪比条件下具有鲁棒性。该算法计算代价小,简单易实现,适用于嵌入式语音识别系统。     

基于奇异谱分析和维纳滤波的语音去噪算法

针对维纳滤波算法对非平稳语音信号去噪存在的信号失真、信噪比(SNR)不高的问题,提出了一种奇异谱分析(SSA)和维纳滤波(WF)相结合的语音去噪算法SSA-WF。通过奇异谱分析将非线性、非平稳的语音信号初步去噪,提高含噪语音的信噪比以获取尽可能平稳的语音,并将其作为维纳滤波的输入,以剔除其中仍存在的高频噪声,最终获取纯净的去噪语音。在不同强度的背景噪声下进行仿真实验,结果表明SSA-WF算法在SNR和均方根误差(RMSE)等方面都要优于传统的语音去噪算法,能够有效去除背景噪声,降低有用信号的失真,适用于非线性、非平稳语音信号的去噪。     

IPP在基于小波域Wiener滤波的图像去噪中的应用

为了提高工业在线检测中图像处理的速度,使用英特尔集成性能原件(Intel IPP)作为编程工具,在小波域结合Wie-ner滤波进行图像去噪。对IPP作了简介,并阐述了小波域Wiener滤波原理和ThWiener算法。实验表明,在VC 平台下使用IPP对含噪图像进行小波域Wiener滤波具有良好的去噪效果,并且计算速度快,效率高,在实时图像处理中有较高的应用价值。     

Noise variance estimation based on dual-channel phase difference for speech enhancement

In this paper, we propose a method for estimating a signal-to-noise ratio (SNR) in order to improve the performance of a dual-microphone speech enhancement algorithm. The proposed method is able to reliably estimate both a priori and a posteriori SNRs by exploring a direction-of-arrival (DOA)-based local SNR that is defined by using spatial cues obtained from dual-microphone signals. The estimated a priori and a posteriori SNRs are then incorporated into a Wiener filter. Consequently, it is shown from an objective perceptual evaluation of speech quality (PESQ) comparison and a subjective listening test that a speech enhancement algorithm employing the proposed SNR estimate outperforms those using conventional single- or dual-microphone speech enhancement algorithms such as the Wiener filter, beamformer, or phase error-based filter under different noise conditions ranging from 0 to 20 dB.     

一种听觉掩蔽效应和维纳滤波的语音增强算法

将非平稳噪声估计算法以及基于听觉掩蔽效应得到的噪声被掩蔽概率应用于维纳滤波语音增强中,提出了一种听觉掩蔽效应和维纳滤波的语音增强方法。几种噪声背景下对语音增强的客观测试表明,提出的算法相比较于传统的维纳滤波语音增强算法而言不但可以提高语音信噪比,而且可以明显减少语音失真。     

Speech transmission with COFDM based on different discrete transforms

Recently, the multimedia and cellular technologies have spread dramatically. Therefore, the demand for digital information has increased. Speech compression is one of the most effective forms of communication. This paper presents three approaches for the transmission of compressed speech signals over convolutional Coded Orthogonal Frequency Division Multiplexing (COFDM) system with a chaotic interleavering technique. The speech signal has is compressed using the Set Partitioning In Hierarchical trees (SPIHT) algorithm, which is an improved version of EZW and which is characterized by a simple and effective method for further compression. For mitigation of the fading due to multipath wireless channels, this paper proposes a COFDM system based on fractional Fourier transform (FrFT), a COFDM system based on discrete Cosine transform (DCT), and a COFDM system based on discrete wavelet transform (DWT). The FrFT has the ability of solving the frequency offset problem, which causes the received frequency-domain sub-carriers to be shifted, and therefore, the orthogonality between subcarriers deteriorates even with equalization. The DCT has an advantage of increased computational speed as only real calculations are required. The DWT is spectrally efficient since it does not utilize cyclic prefix (CP). These systems have been designed under the assumption that corruptive background noises are absent. Therefore, denoising techniques, namely wavelet denoising and Wiener filtering methods are suggested at the receiver to achieve enhancement in the speech quality. The simulation experiments shows that the proposed COFDM–DWT with Wiener filtering at the receiver has a better trade-off between BER, spectral efficiency and signal distortion. Hence, the BER performance is improved with small bandwidth occupancy. Moreover, due to the denoising stage, the speech quality is improved to achieve good intelligibility.     

脉冲噪声环境下基于卡尔曼滤波的语音增强

语音增强的目标在于从含噪信号中提取纯净语音,纯净语音在某些环境下会被脉冲噪声所污染,但脉冲噪声的时域分布特征却给语音增强带来困难,使传统方法在脉冲噪声环境下难以取得满意效果。为在平稳脉冲噪声环境下进行语音增强,提出了一种新方法。该方法通过计算确定脉冲噪声样本的能量与含噪信号样本的能量之比最大的频段,利用该频段能量分布情况逐帧判别语音信号是否被脉冲噪声所污染。进一步地,该方法只在被脉冲噪声污染的帧应用卡尔曼滤波算法去噪,并改进了传统算法执行时的自回归(AR)模型参数估计过程。实验中,采用白色脉冲噪声以及有色脉冲噪声污染语音信号,并对低输入信噪比的信号进行语音增强,结果表明所提出的算法能显著地改善信噪比和抑制脉冲噪声。     

Speech enhancement using Teager energy operated ERB-like perceptual wavelet packet decomposition

In recent past, wavelet packet (WP) based speech enhancement techniques have been gaining popularity due to their inherent nature of noise minimization. WP based techniques appeared as more robust and efficient than short-time Fourier transform based methods. In the present work, a speech enhancement method using Teager energy operated equal rectangular bandwidth (ERB)-like WP decomposition has been proposed. Twenty four sub-band perceptual wavelet packet decomposition (PWPD) structure is implemented according to the auditory ERB scale. ERB scale based decomposition structure is used because the central frequency of the ERB scale distribution is similar to the frequency response of the human cochlea. Teager energy operator is applied to estimate the threshold value for the PWPD coefficients. Lastly, Wiener filtering is applied to remove the low frequency noise before final reconstruction stage. The proposed method has been applied to evaluate the Hindi sentences database, corrupted with six noise conditions. The proposed method’s performance is analysed with respect to several speech quality parameters and output signal to noise ratio levels. Performance indicates that the proposed technique outperforms some traditional speech enhancement algorithms at all SNR levels.     

基于语音信号稀疏性的FDICA初始化和后处理方法

目前解决语音信号盲源分离(Blind source separation,BSS)的两大类方法分别为频域独立成分分析(Frequency domain independent component analysis,FDICA)和基于稀疏性的时频掩蔽(Time frequency masking,TF masking).为此将两类方法优点相结合,利用TF masking方法的结果,对FDICA做初始化,在加快FDICA收敛速度的同时也避免了次序不确定性问题.此外还提出了一种新的基于语音稀疏性FDICA的BSS后处理方法:基于局部最小比例控制(Local minimum ratio controlled,LMRC)谱减法,比常规的TF masking、维纳滤波等后处理方法,能够更有效地控制音乐噪声,提高分离性能.合成数据和实际采集数据的实验结果验证了所提方法的有效性.     

两种改进的图像复原算法在COSM中的应用*

对模糊的成像结果进行图像复原,采用高斯函数作为点扩展函数,应用于三维逆滤波和维纳滤波算法中,改进了这两种算法。实验结果证明,改进后的维纳滤波算法复原三维序列的结果比改进后逆滤波算法得到的结果更好。此外,提出三维维纳增量滤波算法以及加快其收敛速度的方法。