一种基于后验信噪比频域语音增强方法的研究

谱减法是常用的单通道语音降噪方法,传统谱减法在抑制背景噪声的同时引入了“音乐噪声”,影响听觉效果。为了抑制音乐噪声,提出了一种基于后验信噪比的频域语音增强新方法,当后验信噪比较高时,采用基于后验信噪比的谱减法增强语音信号;当后验信噪比较低时,采用基于后验信噪比的谱衰减方法对含噪语音信号谱线进行衰减,达到语音增强的目的。仿真结果表明,基于后验信噪比的频域语音增强法具有较好的背景噪声和音乐噪声抑制效果,并保持了较好语音可懂度。     

基于频谱减法的语音去噪算法研究

语音增强技术是音频信号处理中的重要部分,频谱减法是目前在语音增强技术中最常用的方法之一。针对传统频谱减法会产生音乐噪声并无法消除音乐噪声的不足之处及高频噪声干扰比较严重的情况下频谱减法效果差的情况,采用了在频谱减法之后进行LMS滤波以降低音乐噪声对语音质量的影响和低通滤波以滤除脉冲干扰。根据仿真结果表明,改进扩展频谱减法能够有效降低音乐噪声和尖锐的高频兹兹声,从而提高信噪比,达到语音增强的目的。     

一种引入延迟的语音增强算法

针对传统语音增强算法中,只采用当前帧和当前帧以前的信息对当前帧语音谱进行估计而造成变电平噪声和音乐噪声的问题,采用一种改进的引入延迟的语音增强算法。通过引入延迟,可以在对当前帧语音谱进行估计时使用当前帧以后帧的信息,在噪声估计中采用类似路径搜索的双向搜索方法消除变电平噪声的影响,在先验信噪比估计中采用改进的非因果先验信噪比估计算法,消除低信噪比平滑不足带来的音乐噪声,在此基础上构建了一个完整的语音增强算法。实验结果表明,该算法基本不受变电平噪声的影响,而且音乐噪声和残留背景噪声都得到了很好的抑制。     

基于两步噪声消除技术与高斯统计模型的语音增强算法

针对语音增强技术中先验信噪比参数的估计问题,本文通过结合两步噪声消除技术以及语音与噪声分量的高斯统计模型,在频率域中提出了一种新的先验信噪比估计算法。该算法基于直接判决方法的输出结果,利用最小均方误差估计理论直接计算当前帧纯净语音分量的谱能量,以获取带噪语音的先验信噪比估计。算法在保留两步噪声消除算法优点的基础上,无需语音增强系统中增益因子的任何先验条件,且在有效消除背景噪声的同时能够最大程度地抑制输出语音中音乐噪声的生成。多种噪声背景下的仿真结果表明:相对于经典的直接判决方法和新近的两步噪声消除算法,基于本文先验信噪比估计方案的语音增强系统在主观与客观评价标准下都具有更加优良的语音增强效果。      

基于参数估计和感知提升的语音增强降噪算法

为了提高单通道语音增强降噪算法的整体质量,该文从噪声消除和语音感知两个角度出发对传统语音增强算法进行改进,通过引入多种处理手段来达到最佳优化效果。首先在参数估计方面,把基于弱语音出现的平滑算法加入到基于固定先验信噪比的软判决方法中来解决噪声谱过估计问题,并根据语音帧存在概率动态调整平滑因子,从而提高先验信噪比的跟踪效果。其次在语音质量感知提升方面,采用谐波恢复的方法重建语音段的高频谐波分量,并采用相位补偿和增益平滑的方法消除静默段和语音段的音乐噪声。实验结果表明,相比传统算法,该文算法通过引入参数估计改进模块和感知质量提升模块,在消噪效果和语音质量两方面均得到了较大的提高,并适用于多类噪声环境和信噪比条件。     

基于匹配追踪与子空间联合的语音增强

提出了基于匹配追踪和基于子空间相结合的语音增强方法.匹配追踪法具有自适应性,可以从一个过完备字典中找出与信号匹配的原子,重构后得到去除噪声的语音信号.针对匹配追踪法在输入信噪比较低时可能出现信号原子被误判为噪声成分的问题,引入子空间方法.首先分离出噪声子空间,再与信号子空间进行匹配追踪分解.实验结果表明,这种方法对各种信噪比的语音信号都有较好的增强效果,并有效抑制了音乐噪声.     

改进的低时延语音增强算法

针对目前语音增强存在较大时延的问题,提出一种低时延语音增强算法。在传统的先验信噪比估计和噪声估计的基础上,对判决引导算法进行低时延的改进,并提出了一种基于语音存在概率和语音激活检测相结合的噪声估计方法,本文的算法采用对数MMSE估计器结合语音存在概率。采用ITU-T P.826 PESQ、分段信噪比、总信噪比和对数谱失真对该算法进行了测试,并与其他几种算法进行了对比,实验结果表明,该算法有效降低了时延,可以很好的跟踪非平稳噪声,在信噪比较低的情况小可以取得很好的增强效果,且音乐噪声和残留背景噪声也可以得到很好的抑制。     

基于人耳感知掩蔽效应的子空间语音增强算法研究

针对传统子空间语音增强算法中,因语音增强方法中去除噪声而出现的音乐噪声和失真问题,提出了一种人耳感知掩蔽效应的子空间语音增强算法,并结合频域到特征值域的变换,在Bark域内实现人耳的感知掩蔽效应的语音增强。实验结果表明,该算法在白噪声和有色噪声的背景下,与传统子空间语音增强算法相比,不仅提高了语音信号的信噪比,而且减少了语音失真和音乐噪声,提高了增强后语音的听觉质量。     

一种低信噪比语音的增强算法

为改善低信噪比环境下语音的质量,论文提出了一种新的语音增强算法。算法首先根据噪声频谱的高斯统计模型得到用先验信噪比形式表示的噪声频谱估计值,然后利用帧内、帧间平滑算法估计每一个频点的先验信噪比,从而能够更好地跟踪先验信噪比的变化。算法接着引入一种简便的估计语音在每一个频点出现概率的方法,得出一种新的语音增强算法。客观测试和非正式听音测试表明:该算法在几乎不损伤语音清晰度的前提下,能够更好地抑制低信噪比语音增强所产生的音乐噪声,同时使语音信噪比得到了明显提高。     

基于听觉掩蔽效应和噪声估计的语音增强算法

谱减法在增强语音、提高信噪比的同时,残留的音乐噪声较大.在利用听觉掩蔽闻值对谱减系数进行修正的基础上,采用实时噪声估计来减少谱减法噪声估计误差,并对谱减后的语音信号进行感知滤波来进一步抑制残留音乐噪声.实验结果表明,该算法能去除噪声,增强语音,并在不影响信噪比的同时降低语音失真测度值.主观测听表明语音音质有明显提高.     

信噪比对DIMM法测量大气相干长度的影响

在大气相干长度r0的测量中,相同的仪器和数据处理方法,不同的信噪比对实验测得结果有较大影响。通过分析实测中的噪声模型,利用matlab仿真,保持信号光强度不变,在光斑图像上添加高斯噪声,改变图像的信噪比。以不含背景时的光斑图像求得的r0作真值,分别求出不同信噪比下r0对真值的相对偏差。发现当信噪比小于5时,相对偏差大幅增加。在相同信噪比下,采用不同的背景处理方法,发现减去平均背景后,剩下的高斯噪声对质心位置的影响很小。在不同信噪比下,减1倍平均背景及0.6倍背景起伏标准差后相对偏差整体表现最小。     

Feature classification criterion for missing features mask estimation in robust speaker recognition

Currently, many speaker recognition applications must handle speech corrupted by environmental additive noise without having a priori knowledge about the characteristics of noise. Some previous works in speaker recognition have used the missing feature (MF) approach to compensate for noise. In most of those applications, the spectral reliability decision step is performed using the signal to noise ratio (SNR) criterion, which attempts to directly measure the relative signal to noise energy at each frequency. An alternative approach to spectral data reliability has been used with some success in the MF approach to speech recognition. Here, we compare the use of this new criterion with the SNR criterion for MF mask estimation in speaker recognition. The new reliability decision is based on the extraction and analysis of several spectro-temporal features from across the entire speech frame, but not across the time, which highlight the differences between spectral regions dominated by speech and by noise. We call it the feature classification (FC) criterion. It uses several spectral features to establish spectrogram reliability unlike SNR criterion that relies only in one feature: SNR. We evaluated our proposal through speaker verification experiments, in Ahumada speech database corrupted by different types of noise at various SNR levels. Experiments demonstrated that the FC criterion achieves considerably better recognition accuracy than the SNR criterion in the speaker verification tasks tested.     

Comparison of fiber-optic Sagnac and Mach-Zehnder interferometerswith respect to thermal processes in the fiber

The impact of thermal noise in the fiber on the noise power and signal-to-noise ratio (SNR) of fiber-optic sensors is investigated, and sensors based on the Sagnac and the Mach-Zehnder interferometers are compared. The SNR is almost the same for the two interferometers; the Mach-Zehnder interferometer has slightly higher SNR at low frequencies, whereas the Sagnac interferometer may have a few dB higher SNR at high frequencies. It is also shown that thermal noise may dominate over shot noise for realistic system parameters, in which case a Sagnac interferometer has several advantages over a Mach-Zehnder interferometer     

噪声功率不确定下协同频谱感知最优融合算法

研究了噪声功率波动性对单点和协同频谱感知性能地影响。分析了在此环境下,单点基站的检测概率与采样点个数的关系,并得出当信噪比不小于噪声功率波动系数的两倍时,基站仍可正常检测。在协同网络中,给出了此环境下数据融合方法的最优权重设定。理论分析与实验表明,相比于传统方法,文中所提最优数据融合算法在各种信噪比环境下性能均较优,尤其在各基站信噪比差异较大时,性能更为显著。     

SNR wall and cooperative spectrum sensing in cognitive radio under noise uncertainty

Cooperative energy spectrum sensing has been proved effective to detect the spectrum holes in Cognitive Radio (CR). However, its performance may suffer from the noise uncertainty, which is portrayed by the SNR wall in some literatures. In this paper we analyze the spectrum sensing performance under noise uncertainty and find an alternative approach to obtain the SNR wall. Then the average SNR wall is proposed to illustrate the statistically average impact of noise uncertainty. In addition, the cooperative sensing performance under noise uncertainty with AND rule is discussed. Analyses and numerical results show that cooperative sensing can significantly improve the sensing performance under the condition of noise uncertainty.     

采用矩阵累乘的自相关矩阵构造

提出了一种适用于低信噪比情况下提取扩频信号特征参数的算法。该算法通过对被测信号的多次采样、分段累乘,扩大了待分解信号的样本数,降低了噪声的影响,从而能够获得比传统子空间分解算法更好的性能。通过分段累乘构造的自相关矩阵,对其进行特征值分解后,表现出对噪声不敏感的特性,在一定程度上克服了常规方法的噪声敏感缺点。对算法的仿真计算表明,该方法应用在低信噪比的通信环境下,信号特征值不会被噪声湮没,解决了传统子空间方法在低信噪比条件下的分辨率不足的问题。该算法的提出对低信噪比条件下的扩频信号处理和参数检测有重要的工程和实际意义。     

The method of separating harmonic signals from multiplicative and additive noises

This paper focuses on the extraction of a harmonic signal from multiplicative and additive noises. A method is proposed in two stages: (1) to square the original discrete time series, which in- cludes both signals and noises, and form a new time series. By this means, the multiplicative noise is converted to additive noise; and (2) to filter out the noise by using existing noise removal schemes. With a large amount of simulation, experimental results demonstrated the efficiency and effective- ness of this newly developed method in terms of Signal-to-Noise Ratio (SNR) and other criteria. From the experiment, it is also found that: the two kinds of noises affect the SNR differently. In gen- eral, the SNR is not influenced by multiplicative Gaussian noise regardless of its variance. However, if both kinds of noise exist, the SNR decreases with the incensement of the Variance of Additive Noise to Multiplicative Noise Ratio (VAMNR). This analysis is also supported by simulation work.     

Speech enhancement using constrained spectral amplitude subtraction based on noncausal a priori SNR

Two gain forms of spectral amplitude subtraction are derived theoretically without neglecting the correlation of speech and noise spectrum during the period of a fralne. In the implementation, the constrained gain is expressed as a function of noncausal a priori SNR （Signal-to-Noise Ratio）. Noise and noncausal a priori SNR are estimated from the multitaper spectrum of the noisy signal with algorithms modified to be suitable for the multitaper spectruln. Objective evaluations show that in case of white Gaussian noise the proposed method outperforms some methods based on LSA （Log Spectral Amplitude） in terms of MBSD （Modified Bark Spectral Distortion）, segmental SNR and overall SNR, and informal listening tests show that speech reconstructed in this way has little speech distortion and musical noise is nearly inaudible even at low SNR.     

基于自相关功率谱的生命迹象探测算法

在实际雷达生命迹象探测环境中,干扰噪声往往不是理想的高斯白噪声,而是非零均值、且具有相关性的高斯状色噪声。针对互功率谱算法在非理想高斯白噪声背景下提高信噪比能力有限这一问题,通过对其不足与缺陷的分析,提出一种在高斯状色噪声条件下,应用于步进频连续波生命探测雷达,基于自相关功率谱的生命迹象探测算法。该算法利用源目标回波信号的自相关性,对基带回波信号进行自相关处理,以此增强源目标信号功率谱密度,提高信噪比,提升探测性能。通过仿真,证明了该算法提升信噪比能力。同时利用步进频连续波生命探测雷达进行实际测试,统计、分析了探测环境中的噪声模型与数字特征,验证了自相关功率谱在实际探测环境中,具有更强的生命迹象检测性能。