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.     

Robust feature extraction based on an asymmetric level-dependent auditory filterbank and a subband spectrum enhancement technique

In this paper we introduce a robust feature extractor, dubbed as robust compressive gammachirp filterbank cepstral coefficients (RCGCC), based on an asymmetric and level-dependent compressive gammachirp filterbank and a sigmoid shape weighting rule for the enhancement of speech spectra in the auditory domain. The goal of this work is to improve the robustness of speech recognition systems in additive noise and real-time reverberant environments. As a post processing scheme we employ a short-time feature normalization technique called short-time cepstral mean and scale normalization (STCMSN), which, by adjusting the scale and mean of cepstral features, reduces the difference of cepstra between the training and test environments. For performance evaluation, in the context of speech recognition, of the proposed feature extractor we use the standard noisy AURORA-2 connected digit corpus, the meeting recorder digits (MRDs) subset of the AURORA-5 corpus, and the AURORA-4 LVCSR corpus, which represent additive noise, reverberant acoustic conditions and additive noise as well as different microphone channel conditions, respectively. The ETSI advanced front-end (ETSI-AFE), the recently proposed power normalized cepstral coefficients (PNCC), conventional MFCC and PLP features are used for comparison purposes. Experimental speech recognition results demonstrate that the proposed method is robust against both additive and reverberant environments. The proposed method provides comparable results to that of the ETSI-AFE and PNCC on the AURORA-2 as well as AURORA-4 corpora and provides considerable improvements with respect to the other feature extractors on the AURORA-5 corpus.     

基于语音增强失真补偿的抗噪声语音识别技术

本文提出了一种基于语音增强失真补偿的抗噪声语音识别算法。在前端,语音增强有效地抑制背景噪声;语音增强带来的频谱失真和剩余噪声是对语音识别不利的因素,其影响将通过识别阶段的并行模型合并或特征提取阶段的倒谱均值归一化得到补偿。实验结果表明,此算法能够在非常宽的信噪比范围内显著的提高语音识别系统在噪声环境下的识别精度,在低信噪比情况下的效果尤其明显,如对-5dB的白噪声,相对于基线识别器,该算法可使误识率下降67.4%。     

Spectral domain texture analysis for speech enhancement

A fast algorithm which aims at performing texture analysis of time-frequency images for denoising purposes is described in this paper. Time-frequency images are built using the peaks of the amplitude spectrum computed on a noisy speech signal. Using texture analysis, we can look at the spectral 2D information on a large scale, thus allowing the correction of spectral continuity by restoring peaks corrupted by noise which can appear as missing or modified. The algorithm has been used in preprocessors of speech processing systems. In fact, we report interesting results obtained with this algorithm in speech enhancement and HMM speech recognition tasks, especially for noise types which are quite difficult to treat with conventional algorithms, such as micro-interruptions or bursts of tonal noise at random frequencies.     

基于增益字典查询的语音增强算法

对于基于统计模型的语音增强算法,不同分布模型对应于不同的增益函数,由于语音信号的不确定性,没有一种分布函数能准确对语音和噪声谱的分布建模,因此任何一种固定的统计模型均会存在一定的误差。所以提出一种增益字典查询的语音增强算法,该算法通过采用对数谱失真准则对一个语音噪声库进行增益的训练,得到一个增益的字典,其中输入为先验信噪比和后验信噪比的估计值。最后采用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,并与基于高斯分布模型、拉普拉斯分布模型的算法进行了对比。实验结果表明,该算法无论在非平稳噪声还是平稳噪声环境下都比其他几种算法增强效果好,且音乐噪声和残留背景噪声也可以得到很好的抑制。     

基于时频结合的背景噪声下语音增强方法

本文在研究基于改进谱减法的基础之上,提出了在频域利用谱减法去除宽带加性噪声,在时域利用短时过零率和短时能量组合而成的加权函数去除谱减法后产生的“音乐噪声”的方法．实验表明：这种时频结合的语音增强方法对背景噪声下的语音质量的增强效果明显。     

抑制坦克强背景噪声的改进谱减法研究

谱减法是处理宽带噪声较为传统和有效的方法,它运算量较小,容易实时处理,增强效果也较好。根据经典谱减法及其各种改进形式的基本原理,提出一种新的改进谱减法语音增强算法。根据语音和噪声各自的特性,对带噪语音进行时域平滑和频谱统计加权处理。对该算法进行客观和主观测试表明：相对于传统的谱减法,该算法能更好地抑制背景噪声和音乐噪声,同时也较好地保持了语音的可懂度和自然度。     

稀疏低秩模型下的单通道自学习语音增强算法

针对现有基于字典学习的增强算法依赖先验信息的问题,基于矩阵的稀疏低秩分解提出一种无监督的单通道语音增强算法。该算法首先通过稀疏低秩分解将带噪语音的幅度谱分解为低秩、稀疏和噪声三部分,然后通过对低秩部分进行自学习构建出噪声字典,最后利用所得噪声字典和乘性迭代准则于低秩和稀疏部分中分离出纯净语音。相较于其他基于字典学习的语音增强算法,本文所提算法无需语音或噪声的先验信息,因而更加方便和实用。实验结果显示,本文算法能够在保留语音谐波结构的同时有效抑制噪声,增强效果明显优于鲁棒主成分分析和多带谱减法。     

稀疏低秩模型及相位谱补偿的语音增强算法

针对现有的语音增强算法存在增强效果差、语音信号失真等问题,提出了稀疏低秩模型及改进型相位谱补偿的语音增强算法。首先,用稀疏低秩模型处理含噪语音的幅度谱,得到分离后的语音。接着,用归一化最小均方自适应滤波算法优化相位谱补偿算法的补偿因子。然后,对稀疏低秩分离后的语音进行改进型相位谱补偿处理,得到最终增强的语音。最后,对增强后的语音进行感知语音质量评价分析及频谱分析。实验结果表明,该方法能够有效地去除噪声,并且在低信噪比的情况下,可以保持语音的清晰度。     

一种改进的维纳滤波语音增强算法

提出了一种改进的语音增强算法,该算法以基于先验信噪比估计的维纳滤波法为基础。首先通过计算无声段的统计平均得到初始噪声功率谱;其次,计算语音段间带噪语音功率谱,并平滑处理初始噪声功率谱和带噪语音功率谱,更新了噪声功率谱;最后,考虑了某频率点处噪声急剧增大的情况,通过计算带噪语音功率谱与噪声功率谱的比值,自适应地调整噪声功率谱。将该算法与其他基于短时谱估计的语音增强算法进行了对比实验,实验结果表明：该算法能有效地减少残留噪声和语音畸变,提高语音可懂度。     

基于语音存在概率和听觉掩蔽特性的语音增强算法

低信噪比下,谱减语音增强法中一直存在的去噪度、残留的音乐噪声和语音畸变度三者间均衡这一关键问题显得尤为突出。为降低噪声对语音通信的干扰,提出了一种适于低信噪比下的语音增强算法。在传统的谱减法基础上,根据噪声的听觉掩蔽阈值自适应调整减参数,利用语音存在概率,对语音、噪声信号估计,避免低信噪比下端点检测(VAD)的不准确,有更强的鲁棒性。对算法进行了客观和主观测试,结果表明：相对于传统的谱减法,在几乎不损伤语音清晰度的前提下该算法能更好地抑制残留噪声和背景噪声,特别是对低信噪比和非平稳噪声干扰的语音信号,效果更加明显。     

面罩语音质量评价算法适用性研究

针对语音编码的音质评价算法性能已十分明确,但对于面罩语音不一定适用。讨论了语音质量评价算法对空气语音与面罩语音在不同噪声环境下的适用性。采用主观意见得分和三种客观评价测度对多种信噪比的带噪语音和增强语音进行评价,包括分段信噪比、改进的巴克谱失真（MBSD）和语音感知质量评价（PESQ）,根据与主观评价的一致性判断客观评价方法的适用性。增强算法采用维纳滤波法和对数谱最小均方误差法（LSA-MMSE）,噪声采用粉红噪声、海浪噪声。仿真结果表明,语音质量评价算法的适用性与语音类型、信噪比、背景噪声、增强算法种类有关。粉红噪声环境下,PESQ不适合评价经维纳滤波增强的空气语音;MBSD算法只适用于评价经LSA-MMSE增强的面罩语音。海浪噪声环境下,PESQ适用于评价面罩语音,MBSD不适合评价面罩语音。     

多数据流子带噪声语音识别方法

提出一种噪声下的多数据流子带语音识别方法。传统的子带特征方法虽然能提高噪声下的语音识别性能,但通常会使无噪声情况下的识别性能下降。新方法提取感知线性预测(PLP)特征和子带特征,分别进行识别,然后在识别概率层将两者相结合。通过E-Set在NoiseX92下的白噪声的识别实验表明,新方法不仅具有更好的抗噪性能,而且同时能提高无噪声情况下的识别性能。     

Environment-Optimized Speech Enhancement

In this paper, we present a training-based approach to speech enhancement that exploits the spectral statistical characteristics of clean speech and noise in a specific environment. In contrast to many state-of-the-art approaches, we do not model the probability density function (pdf) of the clean speech and the noise spectra. Instead, subband-individual weighting rules for noisy speech spectral amplitudes are separately trained for speech presence and speech absence from noise recordings in the environment of interest. Weighting rules for a variety of cost functions are given; they are parameterized and stored as a table look-up. The speech enhancement system simply works by computing the weighting rules from the table look-up indexed by the a posteriori signal-to-noise ratio (SNR) and the a priori SNR for each subband computed on a Bark scale. Optimized for an automotive environment, our approach outperforms known-environment-independent-speech enhancement techniques, namely the a priori SNR-driven Wiener filter and the minimum mean square error (MMSE) log-spectral amplitude estimator, both in terms of speech distortion and noise attenuation.     

基于感知掩蔽的重构非负矩阵分解单通道语音增强算法

针对非负矩阵分解（NMF）语音增强算法在低信噪比（SNR）非稳定环境下存在噪声残留的问题,提出一种基于感知掩蔽的重构NMF（PM-RNMF）单通道语音增强算法。首先,将心理声学掩蔽特性应用于NMF语音增强算法中;其次,对不同频率位采用不同的掩蔽阈值,建立自适应感知掩蔽增益函数,通过阈值约束残余噪声能量和语音失真能量;最后,结合语音存在概率（SPP）进行感知增益修正,重构NMF算法,以此建立新的目标函数。仿真结果表明,在不同SNR的3种非稳定噪声环境下,与NMF、重构NMF（RNMF）、感知掩蔽深度神经网络（PM-DNN）算法相比,PM-RNMF算法的感知语音质量评估（PESQ）平均值分别提高了0.767、0.474、0.162,信源失真比（SDR）平均值分别提高了2.785、1.197、0.948。实验结果表明,无论是在低频还是高频PM-RNMF有更好的降噪效果。     

改进相位谱补偿的语音增强方法

针对传统单通道语音增强方法中用带噪语音相位代替纯净语音相位重建时域信号，使得语音主观感知质量改善受限的情况，提出了一种改进相位谱补偿的语音增强算法。该算法提出了基于每帧语音输入信噪比的Sigmoid型相位谱补偿函数，能够根据噪声的变化来灵活地对带噪语音的相位谱进行补偿；结合改进DD的先验信噪比估计与语音存在概率算法（SPP）来估计噪声功率谱；在维纳滤波中结合新的语音存在概率噪声功率谱估计与相位谱补偿来提高语音的增强效果。相比传统相位谱补偿（PSC）算法而言，改进算法可以有效抑制音频信号中的各类噪声，同时增强语音信号感知质量，提升语音的可懂度。     

Robust vocabulary recognition clustering model using an average estimator least mean square filter in noisy environments

Noise estimation and detection algorithms must adapt to a changing environment quickly, so they use a least mean square (LMS) filter. However, there is a downside. An LMS filter is very low, and it consequently lowers speech recognition rates. In order to overcome such a weak point, we propose a method to establish a robust speech recognition clustering model for noisy environments. Since this proposed method allows the cancelation of noise with an average estimator least mean square (AELMS) filter in a noisy environment, a robust speech recognition clustering model can be established. With the AELMS filter, which can preserve source features of speech and decrease the degradation of speech information, noise in a contaminated speech signal gets canceled, and a Gaussian state model is clustered as a method to make noise more robust. By composing a Gaussian clustering model, which is a robust speech recognition clustering model, in a noisy environment, recognition performance was evaluated. The study shows that the signal-to-noise ratio of speech, which was improved by canceling environment noise that kept changing, was enhanced by 2.8 dB on average, and recognition rate improved by 4.1 %.     

Factor Analyzed Subspace Modeling and Selection

We present a novel subspace modeling and selection approach for noisy speech recognition. In subspace modeling, we develop a factor analysis (FA) representation of noisy speech, which is a generalization of a signal subspace (SS) representation. Using FA, noisy speech is represented by the extracted common factors, factor loading matrix, and specific factors. The observation space of noisy speech is accordingly partitioned into a principal subspace, containing speech and noise, and a minor subspace, containing residual speech and residual noise. We minimize the energies of speech distortion in the principal subspace as well as in the minor subspace so as to estimate clean speech with residual information. Importantly, we explore the optimal subspace selection via solving the hypothesis test problems. We test the equivalence of eigenvalues in the minor subspace to select the subspace dimension. To fulfill the FA spirit, we also examine the hypothesis of uncorrelated specific factors/residual speech. The subspace can be partitioned according to a consistent confidence towards rejecting the null hypothesis. Optimal solutions are realized through the likelihood ratio tests, which arrive at the approximated chi-square distributions as test statistics. In the experiments on the Aurora2 database, the FA model significantly outperforms the SS model for speech enhancement and recognition. Subspace selection via testing the correlation of residual speech achieves higher recognition accuracies than that of testing the equivalent eigenvalues in the minor subspace.     

基于子带分解的DFRFT自适应滤波语音增强算法

提出一种改进的语音增强方法,利用子带分解对带噪语音信号进行处理,再在离散分数傅里叶变换(DFRFT)域采用最小均方(LMs)自适应算法进行滤波,对滤波后的子带信号进行DFRFT逆变换,最后利用综合滤波器组合成增强后的语音信号。仿真结果表明,本算法明显提高了收敛速度,减少了计算时间。在主客观评价中均具有较好的语音增强效果。     

用于自动语音识别系统的切换语音功率谱估计算法

针对语音识别系统在噪声环境下不能保持很好鲁棒性的问题,提出了一种切换语音功率谱估计算法。该算法假设语音的幅度谱服从Chi分布,提出了一种改进的基于最小均方误差（MMSE）的语音功率谱估计算法。然后,结合语音存在的概率（SPP）,推导出改进的基于语音存在概率的MMSE估计器。接下来,将改进的MSME估计器与传统的维纳滤波器结合。在噪声干扰比较大时,使用改进的MMSE估计器来估计纯净语音的功率谱,当噪声干扰较小时,改用传统的维纳滤波器以减少计算量,最终得到用于识别系统的切换语音功率谱估计算法。实验结果表明,所提算法相比传统的瑞利分布下的MMSE估计器在各种噪声的情况下识别率平均提高在8个百分点左右,在去除噪声干扰、提高识别系统鲁棒性的同时,减小了语音识别系统的功耗。