基于非齐次隐马尔可夫模型的特定人元音的识别方法

针对特定人汉语元音的语音识别,提出一种基于非齐次隐马尔可夫模型的识别方法.该方法首先提取声道频率响应作为特征参数,然后建立非齐次隐马尔可夫模型来更为精确地刻画真实的语音现象,接着进行语音识别实验,并与齐次隐马尔可夫模型进行比较.实验结果表明该方法可以使特定人的元音的识别率达到98.73%,明显改变了识别系统的性能.该方法具有很好的理论研究前景和实际应用价值.     

基于嵌入式系统的特定说话人语音识别系统的实现

语音识别是实现人机通信的一种重要技术手段.本文基于Samsung公司开发的B lues芯片,应用隐马尔可夫模型(HMM)理论和方法,完成了针对特定人、小词汇量嵌入式语音识别系统开发.文章介绍了语音识别过程的各个环节及其实现方法.     

基于HMM和PNN的语音情感识别研究

语音情感识别是从语音的角度赋予计算机理解情感特征的能力,最终使计算机能像人一样进行自然、亲切和生动的交互。提出了一种融合隐马尔科夫模型（hidden markov model,HMM）和概率神经网络（probabilistic neural network,PNN）的语音情感识别方法。在所设计情感识别系统中,提取出基本的韵律参数和频谱参数,利用PNN处理声学参数的统计特征,利用HMM处理声学参数的时序特征,运用加法规则和乘法规则融合了统计特征和时序特征的识别结果。实验结果显示,所提出的算法在语音情感识别中具有有效的识别能力。     

Using frame correlation algorithm in a duration distribution based hidden Markov model

The assumption of frame independence is a widely known weakness of traditional hidden Markov model (HMM). In this paper, a frame correlation algorithm based on the duration distribution based hidden Markov model (DDBHMM) is proposed. In the algorithm, an AR model is used to depict the low pass effect of vocal tract from which stems the inertia leading to frame correlation. In the preliminary experiment of middle vocabulary speaker dependent isolated word recognition, our frame correlation algorithm outperforms the frame independent one. The average error reduction is about 20% .     

混合抗噪语音识别模型的设计与仿真

为解决语音识别过程中的抗噪声及抗干扰问题,提高系统的识别精度,利用隐马尔可夫模型HMM优异的时序建模能力及小波变换可以对信号进行多尺度分析并有效提取信号的局部信息的特点,建立了混合语音识别模型.考虑到在语音信号识别过程中信号的非平稳性,采用并行的识别方法分别获取分类信息,根据混合模型的识别算法做出识别决策,减小了系统对环境的依赖性,提高了其自适应能力.仿真实验结果表明,混合模型识别结果比单一HMM模型或小波模型识别结果更佳,提高了整体的识别速度和识别率.     

Mandarin Digits Speech Recognition Using Support Vector Machines

A method of applying support vector machine (SVM) in speech recognition was proposed, and a speech recognition system for mandarin digits was built up by SVMs. In the system, vectors were linearly extracted from speech feature sequence to make up time-aligned input patterns for SVM, and the decisions of several 2-class SVM classifiers were employed for constructing an N-class classifier. Four kinds of SVM kernel functions were compared in the experiments of speaker-independent speech recognition of mandarin digits. And the kernel of radial basis function has the highest accurate rate of 99.33 %, which is better than that of the baseline system based on hidden Markov models (HMM) (97.08%). And the experiments also show that SVM can outperform HMM especially when the samples for learning were very limited.     

隐马尔柯夫模型语声识别参数估计的研究

本文讨论了两种新的隐 Markov 模型(HMM)参数估计的方法,分别称为最大互信息估计(MMIE)和最大检测函数估计(MDE),并提出了一种改进的 HMM参数估计算法.     

后验概率图与补白模型二次融合的关键词识别

使用全连接神经网络结合Softmax分类器对汉语的408个音节建立音节分类器,利用等长处理后的特征向量训练Softmax分类器,将Softmax分类器输出概率作为后验概率图,与隐马尔科夫补白模型（HMM/Filler）进行第一次融合,得到子后验概率图隐马尔科夫模型（Posteriorgram-HMM）. 针对关键词训练样本较少的问题,将标注样本进行强制切分,得到HMM每个状态上的训练数据. 将隐马尔科夫最大后验概率基线模型（HMM-MAP）与Posteriorgram-HMM进行第二次融合,提出最大后验概率图隐马尔科夫模型（Posteriorgram-HMM-MAP）. 在数据集上训练模型后,使用测试数据对其进行测试. 结果表明：Posteriorgram-HMM-MAP的综合识别率相比Posteriorgram-HMM提升了3.55%,相比HMM/Filler提升了10.29%.     

基于HMM/ANN混合模型的带噪语音识别

对于含噪语音信号的有效特征提取是语音识别至关重要的一步.该文提出了利用小波调制尺度对语音进行特征提取,结合隐马尔可夫和人工神经网络混合模型进行识别的方法,可进一步反映语音信号的动态特性、增强抗干扰能力和提高识别率.实验证明,该模型适合于对噪声背景下的语音进行识别,同传统的HMM模型相比,具有更好的抗噪鲁棒性,在信噪比较低情况下,识别率比传统的HMM模型有明显的提高.     

Novel Extended Phonemic Set for Mandarin Continuous Speech Recognition

An extended phonemic set of mandarin from the view of speech recognition is proposed. This set absorbs most principles of some other existing phonemic sets for mandarin, like Worldbet and SAMPA-C, and also takes advantage of some practical experiences from speech recognition research for increasing the discriminability between word models. And the experiments in speaker independent continuous speech recognition show that hidden Markov models defined by this phonemic set have a better performance than those based on initial/final units of mandarin and have a very compact size.     

基于HMM的声调语音模型研究

针对声韵母相同但声调不同的近音字识别问题和声韵母及声调都相同的同音字识别问题,提出在语音模型和语言模型中分别引入声调和字转移概率,以提高近音字和同音字的识别率。首先将声调划分为5种表现形式添加到汉语音节的最后一个音素中构成新音素,使用高斯混合隐马尔科夫模型建模新音素。然后通过统计方法计算特定语境下的字间转移概率。最后使用HTK工具包实现了带声调的语音模型和有字转移概率的语言模型。实验结果证明添加声调可以提高近音字的识别率,使用特定语境下字间转移概率可以提高同音字的识别率。     

Place recognition based on saliency for topological localization

1 INTRODUCTIONLocalization is fundamental in the domain ofmobile robot navigation. Current localizationmethods based on camera can be mainly classifiedas geometric and topological . Topological methodsuse an adjacent graphin replace of exact coordinateto represent environment . For this reason, theyare dominant inlocalization based on monocular vi-sion[1]. In general , these topological approachesare based on either landmark detection or place rec-ognition.For landmark detection, mobile …     

改进的隐马尔可夫表情识别模型参数优化算法

人脸表情识别是人工智能领域中极富挑战性的课题,针对表情识别中存在的识别率低与计算量大的问题,提出了一种新的改进的隐马尔可夫表情识别模型参数优化的算法.先采用新的初始参数优化模型,然后利用Baum-Welch算法进行重估计,从而建立新的HMM人脸表情模型.实验结果表明,新模型明显提高了人脸表情的识别率并降低了计算量.     

非特定人语音指令识别系统的研究与实现

为了研究语音识别技术的实用化问题.采用离散隐马尔可夫模型方法,基于单片微控制器加数字信号处理器(MCU DSP)构架,实现了一个非特定人语音指令识别系统.该系统适应性强,每个指令识别时间在1s内,准确率达到96%.     

基于瑞利混合隐马尔科夫模型的语音幅度谱分布估计

针对语音信号处理中语音短时幅度谱分布模型过于单一的问题,提出了一种基于隐马尔科夫模型的语音幅度谱分布估计算法。该算法利用瑞利混合模型作为语音幅度谱分布,采用隐马尔科夫模型将语音分成不同的状态,在每一状态中有一组瑞利混合模型参数与之相对应,通过把语音信号分成不同的状态对语音进行分类,为语音短时谱幅度建立更为准确的模型。     

数字离散信号滤器HMM综合平均法的研究

隐Ｍａｒｋｏｖ模型是在语音识别中得到了成功应用的一种统计建模工具。该研究首次将ＨＭＭ技术引入数字的离散信号滤噪研究并提出了ＨＭＭ综合平均法,经对加噪谐信号,方波信号、斜小信号的计算机重建仿真实验后,表明利用ＨＭＭ进行此项工作有着比较好的效果。     

基于皮肤模板和改进HMM的自动人脸识别系统

介绍一种基于隐马尔可夫模型(hidden Markov module,HMM)的人脸识别系统,该系统对人脸采用普通网络摄像头实时检测,通过皮肤模型进行背景去除,并用改进后的HMM算法进行识别. 实验结果表明,改进后的HMM算法能提高原HMM算法的准确率,采用皮肤模板对检测到的人脸进行精确定位后,进一步提高了识别算法的准确度.     

多讲话者汉语单字识别

本文详细介绍了隐 Markov 模型(HMM)的基本概念和计算概率的前、后向算法,并采用 HMM 作了多讲话者(三人)小词汇量汉语单字识别实验。正确识别率为98%。     

HMM-based noise estimator for speech enhancement

A noise estimator was presented in this paper by modeling the log-power sequence with hidden Markov model （HMM）. The smoothing factor of this estimator was motivated by the speech presence probability at each frequency band. This HMM had a speech state and a nonspeech state, and each state consisted of a unique Gaussian function. The mean of the nonspeech state was the estimation of the noise logarithmic power. To make this estimator run in an on-line manner, an HMM parameter updated method was used based on a first-order recursive process. The noise signal was tracked together with the HMM to be sequentially updated. For the sake of reliability, some constraints were introduced to the HMM. The proposed algorithm was compared with the conventional ones such as minimum statistics （MS） and improved minima controlled recursive averaging （IM- CRA）. The experimental results confirms its promising performance.     

双模型语音识别中的听视觉合成和模型同步异步性实验研究

研究了双模型语音识别系统中前合成和后合成两种听觉视觉合成方法；同时在后合成方法中引入了考虑听觉和视觉同步异步特点的复合模型。仿真实验证明了在声学噪音环境下，后合成方法能够带来比较理想的识别效果；考虑听觉和视觉同步异步性的模型可以有效地提高识别率。