采用特征空间随机映射的鲁棒性语音识别

针对语音识别性能受噪声干扰而显著降低的问题,提出一种采用特征空间随机映射（RP）的鲁棒性语音语音识别方法,并应用于汽车驾驶环境下的语音识别系统。首先,将原始语音特征参数采用随机矩阵线性映射到新的特征空间,使新的特征参数以最大概率保持原始特征之间距离的同时更加接近于高斯分布;然后训练隐马尔可夫模型(HMM),测试时结合多数投票表决方法对初始模式匹配结果进行判决并得到最终语音识别结果。采用日本情报处理学会车载环境下语音识别数据库CENSREC-2进行实验分析,结果表明,随机映射特征使得汽车驾驶环境下的语音识别性能有了很大改善。     

基于最大似然线性回归矩阵的说话人识别算法研究

研究了将自适应领域的最大似然线性回归(Maximum likelihood linear regression, MLLR)变换矩阵作为特征进行文本无关的说话人识别算法. 本文引入了基于统一背景模型的MLLRSV-SVM说话人识别算法, 并在此基础上进行高层音素聚类以进一步提高识别性能. 在采用多种信道补偿技术后, 在NIST SRE 2006年1训练语段-1测试语段同信道和跨信道数据库上, 基于MLLR特征的系统与其他最好的系统性能接近并有很强的互补性, 经过简单线性融合可以极大提高识别性能.     

双模态车载语音控制仿真系统的设计与实现

针对音、视频双模态语音识别能有效地提高噪声环境下的识别率的特性,本文设计了车载语音控制指令识别实验系统。该系统模拟车载环境,把说话时的视频信息融入到语音识别系统中,系统分为模型训练、离线识别和在线识别3部分。在线识别全程采用语音作为人机交互手段,并具备用户自适应的功能。离线识别部分将系统产生的数据分层次进行统计,非常适合进行双模态语音识别算法研究。     

实用环境语音识别鲁棒性技术研究与展望

语音识别系统在实用环境中的鲁棒性是语音识别技术实用化的关键问题。鲁棒性研究的核心问题是如何解决实用环境语音特征和模型与干净环境语音识别系统的失配问题,这涉及到噪声补偿、信道适应、说话人自适应等关键技术。文章综述了语音识别鲁棒性技术研究的主要方法、原理及研究现状,分析了实用环境语音识别中声学模型和语言模型的适应技术,并展望了近期语音识别实用化技术发展的研究方向。     

语音识别中基于i-vector的说话人归一化研究

i-vector是反映说话人声学差异的一种重要特征,在目前的说话人识别和说话人验证中显示了有效性。将i-vector应用于语音识别中的说话人的声学特征归一化,对训练数据提取i-vector并利用LBG算法进行无监督聚类．然后对各类分别训练最大似然线性变换并使用说话人自适应训练来实现说话人的归一化。将变换后的特征用于训练和识别．实验表明该方法能够提高语音识别的性能。     

一种新的基于LBG和DTW的模板训练算法

提出了一种新的基于LBG和DTW结合的模板训练算法,包括模板训练、初始模板设置、空子集处理三个部分,能够完整、有效地解决语音识别中模板训练的问题。该算法实现了语音信号特征矩阵的聚类及其质心的生成,使孤立词语音识别系统更好地适用于非特定人的情况,提高了系统对训练集外说话人语音的正确识别率。设计、实现了一个识别系统,模板训练中较快的收敛速度和系统较高的识别率验证了算法的优良性能。     

基于自适应补偿的文本无关说话人识别

基于高斯混合模型的文本无关说话人识别系统通常采用最大似然算法.在纯净语音环境下,基于这种算法的说话人识别系统具有较好的性能.当系统的训练环境和测试环境失配时,这种算法的误识率急剧上升.针对帧似然概率的统计特性,提出了一种新的非线性补偿方法--自适应得分补偿法.通过对帧似然概率归一化、帧均匀化和重新排序赋值等系列补偿措施,改善了原算法的识别性能.实验结果表明,新的补偿方法能够降低误识率,在开集中乎均可达20%,闭集中平均可达50%.     

话者识别的信道补偿

在文本无关的说话人识别中,训练与测试语音中信道环境的差异是影响其性能最重要的因素.近年来,利用因子分析对信道建模成为说话人识别领域的重要方法,大大降低了说话人确认的错误率,但运算复杂度限制了实时的应用.本文介绍了一种简化的因子分析方法:首先在混合高斯模型的模型域训练信道空间,然后在特征域进行信道补偿,得到的新特征可用于各种系统.在NIST2006的数据库上,利用本文的方法相对基线系统在等错误率上有31%的降低.     

一种基于概率模型的特征补偿算法

本文提出了一种基于概率模型的特征补偿算法.该方法基于语音和噪声的先验概率密度,在倒谱域对语音特征参数进行最小均方误差预测(MMSE),提高识别精度.实验结果表明,本文方法能有效提高噪声环境下的中文连续语音识别的正确率.     

跨库语音情感识别研究进展

语音情感识别在人机交互过程中发挥极为重要的作用, 近年来备受关注. 目前, 大多数的语音情感识别方法主要在单一情感数据库上进行训练和测试 . 然而, 在实际应用中训练集和测试集可能来自不同的情感数据库. 由于这种不同情感数据库的分布存在巨大差异性, 导致大多数的语音情感识别方法取得的跨库识别性能不尽人意. 为此, 近年来不少研究者开始聚焦跨库语音情感识别方法的研究. 本文系统性综述了近年来跨库语音情感识别方法的研究现状与进展, 尤其对新发展起来的深度学习技术在跨库语音情感识别中的应用进行了重点分析与归纳. 首先, 介绍了语音情感识别中常用的情感数据库, 然后结合深度学习技术, 从监督、无监督和半监督学习角度出发, 总结和比较了现有基于手工特征和深度特征的跨库语音情感识别方法的研究进展情况, 最后对当前跨库语音情感识别领域存在的挑战和机遇进行了讨论与展望.     

An Environment-Compensated Minimum Classification Error Training Approach Based on Stochastic Vector Mapping

A conventional feature compensation module for robust automatic speech recognition is usually designed separately from the training of hidden Markov model (HMM) parameters of the recognizer, albeit a maximum-likelihood (ML) criterion might be used in both designs. In this paper, we present an environment-compensated minimum classification error (MCE) training approach for the joint design of the feature compensation module and the recognizer itself. The feature compensation module is based on a stochastic vector mapping function whose parameters have to be learned from stereo data in a previous approach called SPLICE. In our proposed MCE joint design approach, by initializing the parameters with an approximate ML training procedure, the requirement of stereo data can be removed. By evaluating the proposed approach on Aurora2 connected digits database, a digit recognition error rate, averaged on all three test sets, of 5.66% is achieved for multicondition training. In comparison with the performance achieved by the baseline system using ETSI advanced front-end, our approach achieves an additional overall error rate reduction of 12.4%.     

抗噪声语音识别及语音增强算法的应用

提高语音识别系统的鲁棒性是语音识别技术一个重要的研究课题。语音识别系统往往由于训练环境下的数据和识别环境下的数据不匹配造成系统的识别性能下降，为了让语音识别系统在含噪的环境下获得令人满意的工作性能，该文根据人耳听觉特性提出了一种鲁棒语音特征提取方法。在MFCC特征提取之前先对含噪语音特征进行掩蔽特性处理，同时结合语音增强方法对特征进行处理，最后得到鲁棒语音特征。通过4种不同试验结果分析表明，将这种方法用于抗噪声分析可以提高系统的抗噪声能力；同时这种特征的处理方法对不同噪声在不同信噪比有很好的适应性。     

基于GMM-UBM的声纹识别技术的特征参数研究

声纹识别技术实现的关键点在于从语音信号中提取语音特征参数,此参数具备表征说话人特征的能力。基于GMM-UBM模型,通过Matlab实现文本无关的声纹识别系统,对主流静态特征参数MFCC、LPCC、LPC以及结合动态参数的MFCC,从说话人确认与说话人辨认两种应用角度进行性能比较。在取不同特征参数阶数、不同高斯混合度和使用不同时长的训练语音与测试语音的情况下,从理论识别效果、实际识别效果、识别所用时长、识别时长占比等多个方面进行了分析与研究。最终结果表明:在GMM-UBM模式识别方法下,三种静态特征参数中MFCC绝大多数时候具有最佳识别效果,同时其系统识别耗时最长;识别率与语音特征参数的阶数之间并非单调上升关系。静态参数在结合较佳阶数的动态参数时能够提升识别效果;增加动态参数阶数与提高系统识别效果之间无必然联系。     

Incorporating local information of the acoustic environments to MAP-based feature compensation and acoustic model adaptation

The maximum a posteriori (MAP) criterion is popularly used for feature compensation (FC) and acoustic model adaptation (MA) to reduce the mismatch between training and testing data sets. MAP-based FC and MA require prior densities of mapping function parameters, and designing suitable prior densities plays an important role in obtaining satisfactory performance. In this paper, we propose to use an environment structuring framework to provide suitable prior densities for facilitating MAP-based FC and MA for robust speech recognition. The framework is constructed in a two-stage hierarchical tree structure using environment clustering and partitioning processes. The constructed framework is highly capable of characterizing local information about complex speaker and speaking acoustic conditions. The local information is utilized to specify hyper-parameters in prior densities, which are then used in MAP-based FC and MA to handle the mismatch issue. We evaluated the proposed framework on Aurora-2, a connected digit recognition task, and Aurora-4, a large vocabulary continuous speech recognition (LVCSR) task. On both tasks, experimental results showed that with the prepared environment structuring framework, we could obtain suitable prior densities for enhancing the performance of MAP-based FC and MA.     

基于多码本矢量量化的非限定文本的 联机话者辨认方法

传统的利用话者的一个时期的语音作为训练语音,进行话者码本训练的方法,识别系统往往不够稳定．为了适应话者自身语音的时变性,文中提出了利用话者不同时期的语音进行训练话者的模型,每个话者具有多个码本．这些码本是采用逐渐减小误识率的优化过程得到的．为了补偿不同信道对系统识别性能的影响,文中给出了一种信道补偿方法．同时提出以一帧高能的浊音语音特征代替一个浊音音素的特征,实现了在线浊音特征提取,利用两级矢量量化及码本索引策略减少了４４％的识别计算量．这些方法大大增加了系统的识别速度和鲁棒性．文中比较了用ＰＬＰ分析和ＬＰＣ倒谱分析进行话者辨认的识别结果．     

基于MAP+CMLLR的说话人识别中发声力度问题

为了改善发声力度对说话人识别系统性能的影响,在训练语音存在少量耳语、高喊语音数据的前提下,提出了使用最大后验概率（MAP）和约束最大似然线性回归（CMLLR）相结合的方法来更新说话人模型、投影转换说话人特征。其中,MAP自适应方法用于对正常语音训练的说话人模型进行更新,而CMLLR特征空间投影方法则用来投影转换耳语、高喊测试语音的特征,从而改善训练语音与测试语音的失配问题。实验结果显示,采用MAP+CMLLR方法时,说话人识别系统等错误率（EER）明显降低,与基线系统、最大后验概率（MAP）自适应方法、最大似然线性回归（MLLR）模型投影方法和约束最大似然线性回归（CMLLR）特征空间投影方法相比,MAP+CMLLR方法的平均等错率分别降低了75.3%、3.5%、72%和70.9%。实验结果表明,所提出方法削弱了发声力度对说话人区分性的影响,使说话人识别系统对于发声力度变化更加鲁棒。     

Multi-environment model adaptation based on vector Taylor series for robust speech recognition

In this paper, we propose a multi-environment model adaptation method based on vector Taylor series (VTS) for robust speech recognition. In the training phase, the clean speech is contaminated with noise at different signal-to-noise ratio (SNR) levels to produce several types of noisy training speech and each type is used to obtain a noisy hidden Markov model (HMM) set. In the recognition phase, the HMM set which best matches the testing environment is selected, and further adjusted to reduce the environmental mismatch by the VTS-based model adaptation method. In the proposed method, the VTS approximation based on noisy training speech is given and the testing noise parameters are estimated from the noisy testing speech using the expectation-maximization (EM) algorithm. The experimental results indicate that the proposed multi-environment model adaptation method can significantly improve the performance of speech recognizers and outperforms the traditional model adaptation method and the linear regression-based multi-environment method.     

Speaker recognition under stressed condition

This paper presents the feature analysis and design of compensators for speaker recognition under stressed speech conditions. Any condition that causes a speaker to vary his or her speech production from normal or neutral condition is called stressed speech condition. Stressed speech is induced by emotion, high workload, sleep deprivation, frustration and environmental noise. In stressed condition, the characteristics of speech signal are different from that of normal or neutral condition. Due to changes in speech signal characteristics, performance of the speaker recognition system may degrade under stressed speech conditions. Firstly, six speech features (mel-frequency cepstral coefficients (MFCC), linear prediction (LP) coefficients, linear prediction cepstral coefficients (LPCC), reflection coefficients (RC), arc-sin reflection coefficients (ARC) and log-area ratios (LAR)), which are widely used for speaker recognition, are analyzed for evaluation of their characteristics under stressed condition. Secondly, Vector Quantization (VQ) classifier and Gaussian Mixture Model (GMM) are used to evaluate speaker recognition results with different speech features. This analysis help select the best feature set for speaker recognition under stressed condition. Finally, four VQ based novel compensation techniques are proposed and evaluated for improvement of speaker recognition under stressed condition. The compensation techniques are speaker and stressed information based compensation (SSIC), compensation by removal of stressed vectors (CRSV), cepstral mean normalization (CMN) and combination of MFCC and sinusoidal amplitude (CMSA) features. Speech data from SUSAS database corresponding to four different stressed conditions, Angry, Lombard, Question and Neutral, are used for analysis of speaker recognition under stressed condition.     

Speech recognition with improved support vector machine using dual classifiers and cross fitness validation

In this research, a new speech recognition method based on improved feature extraction and improved support vector machine (ISVM) is developed. A Gaussian filter is used to denoise the input speech signal. The feature extraction method extracts five features such as peak values, Mel frequency cepstral coefficient (MFCC), tri-spectral features, discrete wavelet transform (DWT), and the difference values between the input and the standard signal. Next, these features are scaled using linear identical scaling (LIS) method with the same scaling method and the same scaling factors for each set of features in both training and testing phases. Following this, to accomplish the training process, an ISVM is developed with best fitness validation. The ISVM consists of two stages: (i) linear dual classifier that finds the same class attributes and different class attributes simultaneously and (ii) cross fitness validation (CFV) method to prevent over fitting problem. The proposed speech recognition method offers 98.2% accuracy.