Sparse Representation with Optimized Learned Dictionary for Robust Voice Activity Detection

Traditionally, most of voice activity detection (VAD) methods are based on speech features such as spectrum, temporal energy, and periodicity. The robustness of these features plays a critical role on the performance of VAD. However, since these features are always directly generated from observed signal, the robustness of these features would be significantly degraded in non-stationary noise environments, especially at low level signal-to-noise ratio (SNR) condition. This paper proposes a kind of robust feature for VAD based on sparse representation with an optimized learned dictionary. To do so, a speech dictionary and a noise dictionary are first learned from speech corpus and noise corpus, respectively. Then an optimization algorithm is designed to reduce the mutual coherence between the two learned dictionaries. After that the proposed feature is generated from the optimized dictionary-based sparse representation, and a VAD method is derived from the proposed feature. The proposed method is evaluated over seven types of noise and four types of SNR level, experimental results show that the optimized dictionary is important for enhancing the robustness of the proposed method, and the proposed method performs well under non-stationary noise, especially at low level SNR condition.     

采用复高斯分布模型的两步噪声幅度谱估计算法

噪声幅度谱估计是有效抑制外界噪声干扰、提高语音增强算法整体输出性能的重要环节。但目前针对该问题的研究相对较少,常用的语音激活检测算法只能在语音不存在阶段对噪声信号的幅度谱进行更新或估计,无法适用于更为复杂的非平稳噪声环境。为克服这一问题,本文基于噪声频谱的复高斯分布模型假设,提出了新型的两步噪声幅度谱估计算法。算法首先采用软判决技术计算噪声信号的功率谱,然后再结合复高斯分布条件下信号幅度谱和功率谱之间的数学关系间接地获取噪声幅度谱的估计。文中基于这一结论给出了两种估计算法,并在多种噪声环境下对它们的性能进行了仿真评估,其测试结果有效表明了提出算法优良的估计性能。      

基于LPCC和能量熵的端点检测

为提高语音端点检测系统在低信噪比下检测的准确性,提出了一种基于倒谱特征和谱熵的端点检测算法.首先,根据分析得到待测语音帧的倒谱特征量,然后计算该特征量分别在通过训练得到的语音和噪声的高斯混合模型下的似然概率,通过两者概率的比较作出有声无声初判决;联合能量熵端点检测结果得到最终判决,最后通过Hangover机制最大限度的保护了语音.实验结果表明,此方法改善了能量熵端点检测法在babble噪声下的劣势,且在不同噪声环境下均优于G.729 Annex B的性能.     

融合统计模型与EMD的宽带话音增强方法

提出了一种融合统计模型和经验模态分解(EMD)的宽带话音增强方法。该方法首先用统计模型增强算法消除含噪话音中的主要噪声成分,然后用一种基于活动话音检测(VAD)的EMD增强算法做后处理进一步抑制残留噪声,从而使以上2种方法的优点有效地结合。在ITU-T G.160标准下对算法进行了性能测试,测试结果表明,与经典的统计模型方法相比,在不同强度的背景噪声下,增强话音的信噪比提高都较为明显。同时,在低信噪比情况下,该方法能有效抑制增强话音高频部分的音乐噪声,提高了听觉舒适度。     

一种基于噪声估计的语音激活检测算法

针对当前语音激活检测算法在低信噪比和复杂噪声模型的环境下性能损失的问题,提出了一种基于噪声估计的语音激活检测算法,通过对背景噪声进行自适应估计,得到准确的信噪比门限,同时利用估计背景噪声对短时谱进行白化处理,从而使得谱熵判决准则得以适用于复杂噪声模型的环境。实验证明,算法在低信噪比和复杂噪声模型下性能优于G.729B和AMR中的语音激活检测算法。     

Robust Voice Activity Detection Based on Concept of Modulation Transfer Function in Noisy Reverberant Environments

Voice activity detection (VAD) is used to detect speech and non-speech periods from observed speech signals. It is an important front-end technique for many speech technology applications. Many VAD methods have been proposed. However most of them have been applied under clean or noisy conditions. Only a few methods have been proposed for reverberant conditions, particularly under noisy reverberant conditions. We therefore need to understand the ill effects of noise and reverberation on speech to design an accurate and robust method of VAD under noisy reverberant conditions. The ill effects of noise and reverberation for speech can be regarded as the modulation transfer function (MTF) under noisy and reverberant conditions. Therefore, our study is based on the MTF concept to reduce the ill effects of noise and reverberation on speech, and propose a robust VAD method that we obtained in this study. Noise reduction and dereverberation were first applied to the temporal power envelope of the speech signal to restore the temporal power envelope with this method. Then, power thresholding as a VAD decision was designed based on the restored temporal power envelope. A method of estimating the signal to noise ratio (SNR) was proposed to accurately estimate the SNR in the noise reduction stage. Experiments under both artificial and realistic noisy reverberant conditions were carried out to evaluate the performance of the proposed method of VAD and it was compared with conventional VAD methods. The results revealed that the proposed method significantly outperformed the conventional methods under artificial and realistic noisy reverberant conditions.     

Speech Enhancement Algorithm Based on MMSE Short Time Spectral Amplitude in Whispered Speech

An improved method based on minimum mean square error-short time spectral amplitude （MMSE-STSA） is proposed to cancel background noise in whispered speech. Using the acoustic character of whispered speech, the algorithm can track the change of non-stationary background noise effectively. Compared with original MMSE-STSA algorithm and method in selectable mode Vo-coder （SMV）, the improved algorithm can further suppress the residual noise for low signal-to-noise radio （SNR） and avoid the excessive suppression. Simulations show that under the non-stationary noisy environment, the proposed algorithm can not only get a better performance in enhancement, but also reduce the speech distortion.     

一种基于支持向量机的含噪语音的清/浊/静音分类的新方法

本文将支持向量机(SVM)方法应用于语音信号的清/浊/静音检测中,提出并验证了一种在各种信噪比等级下将语音信号有效地分为清音、浊音和静音三类信号的新型分类算法.首先,在高信噪比情况下,本文采用了G.729B VAD中的四个差分参数作为SVM分类器的输入特征参数,进行了静音分类的对比实验,得到了优于G.729B VAD和BP神经网络传统算法的实验结果,说明引入这种机器学习方法做语音分类是可行的,并分析讨论了在核函数不同的情况下支持向量机在实验中所表现出的性能.其次,又讨论了在低信噪比条件下,如何通过对含噪语音建立统计模型,提取对噪音免疫的统计特征参数,并给出了一种对时变背景噪声自适应的估计方法.最后,通过在不同噪音环境下的对比实验结果,验证了本文所提出的算法在中低信噪比情况下的分类性能要优于其他传统算法.     

采用子带长时信号变化特征的稳健语音活动检测

为提高语音活动检测(VAD)在低信噪比下的准确率,提出了一种基于子带长时信号变化特征的VAD算法.将语音信号转换到频域,并分解为几个不重复的子频带,对这些子带信号分别提取长时信号变化特征,然后采用GMM在线建立语音和非语音模型,以模型的似然比进行VAD判决.实验结果表明,算法在较低的信噪比下能够显著地提高语音活动检测的准确率,且在多种噪声环境和信噪比条件下具有较好的稳健性.应用于语音识别系统的实验表明,该算法能有效提高噪声环境下的语音识别率.     

基于长短时能量均值的活动语音检测算法

为了有效抑制非平稳背景噪音对语音处理系统的严重干扰,提出了一种基于长短时能量均值的活动语音检测算法。该算法基于两个合理的假设,一个是基于语音隐含成分集的稀疏分解,不但能尽可能地深留含噪语音中的语音信息,还能在一定程度上消除非语音类噪音的干扰;另一个是对上述稀疏分解的语音进行重构,该重构信号中语音段的时域能量高于非语音段的时域能量。在上述两个假设的基础上,采用重构信号的时域能量作为音频特征,以当前帧为中心,并将与其相邻的特定数量帧的短时能量均值作为当前帧的得分值;以当前帧及其之前特定数量帧的长时能量均值怍为判决阈值,进而提出了以当前帧的短时能量均值和长时能量均值大小作为判断条件的活动语音检测算法。买验结果显示,该算法能有效地区分低信噪比（平稳噪音和忙平稳噪音）条件下的语音和非语音片段,并且其性能优于基于单Gaussian分布的似然比算法．     

基于AR-HMM在线能量调整的语音增强方法

针对单通道语音增强技术对非平稳噪声的跟踪不准确、噪声抑制效果较差的问题,本文提出一种基于在线能量调整的语音增强方法.该方法以归一化临界带能量为特征,采用高斯混合模型对背景噪声进行分类,利用对应类型噪声的自回归隐马尔可夫模型(Auto-Regressive Hidden Markov Model,AR-HMM)和纯净语音的AR-HMM,在最小均方误差准则下估计语音和噪声的功率谱.考虑到非平稳环境中训练集和测试集的差异性,需在线调整语音模型和噪声模型中的能量,语音模型的能量调整采用迭代的期望最大化算法;噪声模型的能量调整则利用的是模型训练过程中的能量重估方法,并以最小值控制的递归平均算法确定噪声能量调整的初始值.在ITU-T G.160标准下对算法进行性能测试,测试结果表明,本文方法对非平稳噪声的跟踪效果较好,对噪声衰减量较大,收敛时间较短.     

Improving Voice Activity Detection via weighting likelihood and dimension reduction

The performance of the traditional Voice Activity Detection （VAD） algorithms declines sharply in lower Signal-to-Noise Ratio （SNR） environments. In this paper, a feature weighting likelihood method is proposed for noise-robust VAD. The contribution of dynamic features to likelihood score can be increased via the method, which improves consequently the noise robustness of VAD. Divergence based dimension reduction method is proposed for saving computation, which reduces these feature dimensions with smaller divergence value at the cost of degrading the performance a little. Experimental results on Aurora Ⅱ database show that the detection performance in noise environments can remarkably be improved by the proposed method when the model trained in clean data is used to detect speech endpoints. Using weighting likelihood on the dimension-reduced features obtains comparable, even better, performance compared to original full-dimensional feature.     

基于分数阶谱相减的语音增强法

该文提出了基于分数阶谱相减的语音增强法(FSS)。该方法通过对带噪语音信号作分数阶傅里叶变换(FRFT),将得到的分数阶语噪混合谱与估计的分数阶噪声谱相减,最后利用分数阶Fourier反变换获得去噪后的语音信号。理论分析表明,所提方法存在一个最佳分数阶阶数,使得语噪混合信号能在分数阶变换域得到最好的分离,从而有效地提高了增强语音的性能。计算机仿真表明,对于混有加性白噪声的男/女声发音信号,所提方法在信噪比提高量和Itakura距离减少量两个方面都优于传统的谱相减法(SS),并且增强语音中的音乐噪声得到了明显抑制。     

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。     

一种基于噪声快速跟踪的语音增强算法

对解决传统减谱算法残留音乐噪声的问题,现有许多方法都无法达到理想效果。提出一种能在非平稳噪声环境下快速追踪噪声的语音增强方法,采用端点检测优化信噪比,达到较好的语音增强效果。实验表明,相比其他类似方法,在提高实时性、增加信噪比和抑制背景噪声和音乐噪声方面都有更好效果。     

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.     

语音业务中鲁棒性VAD算法合析

采用话音激活检测（Voiced Activity Detection，VAD）术的目的是检测语音通信时是否有话音存在，检测到静音时加以抑制，使其不占用或极少占用信道带宽，检测到话音时才对其进行压缩编码与传输。鲁棒性语音识别系统、数字移动通信和因特网实时语音传输等领域要求在恶劣声学环境条件下进行VAD检测，以节省带宽并抑制噪声，因此VAD技术是目前语音处理领域的重要问题。文中给出的几种最新VAD算法（EZCR—VAD，STAT-VAD和E-VAD）是在低信噪比环境下的话音检测具有很好的鲁棒性的算法。     

Improved Speech Denoising Algorithm Based on Discrete Fractional Fourier Transform

The speech signal and noise signal are the typical non-stationary signals,however the speech signa is short-stationary synchronously.Presently,the denoising methods are always executed in frequency domain due to the short-time stationarity of the speech signal.In this article,an improved speech denoising algorithm based on discrete fractional Fourier transform（DFRFT）is pre sented.This algorithm contains linear optimal filtering and median filtering.The simulation shows that it can easily eliminate the noise compared to Wiener filtering improve the signal to noise ratio（SNR）,and enhance the original speech signal.     

基于动态谱估计的改进谱减语音增强算法

语音增强是语音信号处理的重要课题。根据基于最小值追踪的谱估计方法,提出了一种非平稳噪声环境下快速追踪噪声变化的方法,将其应用到改进后的谱减法中,以提升语音增强的效果。仿真结果表明,改进后的谱减法能有效降低背景噪声,提高输出语音信号的信噪比。     

基于实值离散Gabor变换的联合时频域语音增强

传统变换域语音增强方法对语音做短时平稳性假设,这会造成对语音信号和噪声信号谱估计不准确,从而导致语音失真和残留噪声。本文提出一种从联合时频域进行语音增强的方法,该算法无需对语音做短时平稳假设。算法采用具有最佳能量聚集特性的高斯变换核函数,利用能快速实现的实值离散Gabor变换（RDGT）将语音信号变换到联合时频域,然后利用语音和噪声谱服从高斯分布的假设和无语音概率的思想进行基于最小均方误差的语音对数谱估计,采用改进的最小受控递归平均算法（IMCRA）进行噪声时频谱估计,在得到纯净语音的谱估计后利用实值离散Gabor逆变换获得纯净语音估计。实验表明,该算法相比频域变换算法具有较好的语音去噪度和较低的语音失真度。