基于语音参数的普通话韵母区别特征

现有区别特征理论缺少相应语音声学参数的支持,尝试从声学参数的角度来建立普通话韵母的区别特征。首先根据语音参数距离,使用层次聚类方法生成韵母决策树。在此基础上,按开尾韵和有尾韵的韵母分类方案,分析其聚类逻辑,提出了基于参数的普通话韵母区别特征。在非介音韵母中,区别特征层次首先根据主元音音位的高低区分;而在介音韵母中首先通过介音进行区分。使用支持向量机检测区别特征,韵母由区别特征节点组成的搜索路径所确定。通过在区别特征树上的搜索过程实现韵母的检测分类,其中非介音韵母的平均准确率为93.1%,介音韵母为92%。     

《测量管理体系认证》与《计量检测体系确认》的区别

研究《测量管理体系认证》与《计量检测体系确认》在适用范围、审核依据、证书颁发、作用和发展趋势等方面的区别,以便更好地把握《测量管理体系认证》。     

浅谈气体分析仪和气体检测报警仪的区别

1 问题 某化工厂在制氢设备检修前做动火分析,当安全科长用一台可燃气体检测报警仪测定低温制氢设备内氢气含量时,仪器显示值<4%(实际上,几次测定均在3.0%～3.8%H2之间).     

小波变换在声发射信号特征参数检测中的应用

本文论述了小波变换在声发射信号的事件计数特版在数检测中应用的理论和方法,用小波变换来检测声发射信号的事件计数特征参数,其准确率与域值电平的取值在大小无关,从而可大大提高声发射信号特征参数检测的准确度,该方法可用于旋转机械（特别是大型汽轮发电机组）的碰磨故障和裂纹故障的检测。     

不确定度的评定在热膨胀特征参数检测中的应用

讨论了测量不确定度的来源和在热膨胀特征参数检测中不确定度的影响因素，介绍了热膨胀测量不确定度评定的实例。     

非汉语母语者的普通话元音和辅音感知研究

语音感知研究是语音学主要研究内容之一。为了解学习者对普通话元音和辅音的感知情况,并为汉语教学提供借鉴,拓宽言语学习模型的应用领域,减少教学的盲目性,根据第二语言习得理论模式,采用实验语音学和统计学的方法,设计辨认实验和区分实验,分别对普通话水平处于高级和初级水平的20名维吾尔族大学生元音、塞音、擦音和塞擦音的感知情况进行研究。辨认实验考察学习者对元音和辅音的感知反应时间和感知准确率。在区分实验中,计算元音对的频谱距离,辅音对的频谱距离,以及它们的时长差异,对学习者的区分能力进行分析。实验结果显示高级水平学习者的感知元音和辅音的能力明显高于初级水平学习者。学习者对发音部位靠后的元音反应时间快且感知准确率高,对于发音部位靠前的元音反应时间较慢且感知准确率低,对塞擦音的辨认准确率高,对擦音的辨认准确度最低。元音对的频谱距离和辅音对的时长差异会影响到学习者的区分能力,但辅音对的频谱距离和区分情况不存在密切联系。     

“测量管理体系认证”与“计量检测体系确认”的区别

讨论"测量管理体系认证"与"计量检测体系确认"在适用范围、审核依据、证书颁发、作用和发展趋势等方面的区别,以便更好地把握"测量管理体系认证"。     

三坐标机测量形面特征参数补偿方法探讨

三坐标机测量形面特征参数过程中,往往要求测头在某一给定的位置上进行测量.当测头运动方向受到被测形面几何形状的阻挡时,其实际到达的位置偏离于程序给定的位置.这种情况下,测量的结果需要通过补偿进行修正.介绍了两种补偿此测量误差的方法测头半径补偿法、公差带控制法,并分析了这两种补偿方法的特点和适用场合.     

超声功率谱特征参数在复合材料孑L隙率检测中的应用

基于材料超声检测系统模型及功率谱分析,考察了碳纤维增强环氧树脂基（CFRP）复合材料的孔隙率（0．03％～2．3％）与主频、中心频率、加权功率谱频率、有限频带能量（ELFB）及主频对应的能量（EPF）等参数之间的相关性。结果表明：孔隙率在0．03％～0．67％范围内时,ELFB与孔隙率呈线性递减关系;在0．67％～2．3％范围内,两者呈二次曲线递减关系;EPF与孔隙率的关系在0．64％左右两侧分别呈现出线性和二次曲线递减趋势。晶耶和踟能有效表征CFRP的孔隙率。     

浅谈测量不确定度与测量误差的区别

测量不确定度与测量误差是计量科学的基本概念,它们关系着测量结果的可靠程度和量值传递的准确。文章介绍了测量不确定度与测量误差的定义和联系,分析了其区别,并说明表明测量结果的质量时,为什么要用测量不确定度,而不用测量误差。     

Fuzzy clustering based parameter clustering method in mandarin speech recognition

Abstract

In a large vocabulary continuous speech recognition system, to efficiently decrease parameter size and improve the robustness of parameter training, a parameter clustering method by fuzzy clustering is proposed. Based on the structure of the phonetic decision tree, leaf nodes are used for Gaussian clustering and root nodes or shallow leaf nodes are used for covariance sharing. Experimental results show that when the number of Gaussians is reduced by 50%, recognition accuracy only decreases by 0.55%. By combining fuzzy covariance sharing, a total of 4.16% in recognition increase is achieved over the conventional system with approximately the same parameter size.     

基于卷积盲源分离的噪声鲁棒性语音识别的研究

研究了一种基于卷积盲分离算法与MFCC（Mel-Frequency Cepstral Coefficient）特征相结合的噪声鲁棒语音识别方法。该方法在预处理阶段,首先计算预白化观测数据的多阶自相关协方差矩阵,以获得多时延处理的二阶解相关统计信息。然后利用得到的二阶统计信息构建两个对称正定矩阵,通过Cholesky因式分解等一系列变换获得唯一存在的矩阵,根据此矩阵估算语音信号并提取MFCC特征用于后续识别。实验结果表明,在低信噪比条件下,该方法对于数字语音的识别性能优于基本的MFCC识别器和文献中已有的卷积分离算法。     

分段语音时长规整算法

一般的同步叠加算法在进行语音时长规整时,当压扩程度大且语音采样率低时,所得合成信号的语音质量会显著下降。其原因在于同步叠加算法忽略了语音本身的感知重要部分,过度压扩会损害语音的感知效果。针对此现象文章提出一种先根据频谱变化快慢和能量大小将语音划分为感知敏感,非敏感和次敏感部分,对各部分采用不同压扩比进行同步叠加的分段时长规整算法,希望能够提高合成语音质量。实验证明该算法在压扩程度高、低采样率时对语音质量有显著改善。     

Hidden Markov model-based approach for generation of Pitman shorthand language symbols for consonants and vowels from spoken english

Pitman shorthand language (PSL) is a widely practised medium for transcribing/recording speech to text (StT) in English. This recording medium continues to exist in spite of considerable development in speech processing systems (SPS), because of its ability to record spoken/dictated text at high speeds of more than 120 words per minute. Hence, scope exists for exploiting this potential of PSL in present SPS. In this paper, an approach for feature extraction using Mel frequency cepstral coefficients (MFCC) and classification using hidden Markov models (HMM) for generating strokes comprising consonants and vowels (CV) in the process of production of Pitman shorthand language from spoken English is proposed. The proposed method is tested on a large number of samples, drawn from different speakers and the results are encouraging. The work is useful in total automation of PSL processing.     

WDRC释放时间对普通话言语理解度的影响研究

宽动态范围压缩算法作为助听器非线性听力补偿的核心算法,其释放时间常数的设定可影响言语理解度。根据普通话的语音特点,将宽动态范围压缩算法按频率范围划分为低频区间(     

一种新的汉语连续语音声调评测算法

提出一种新的连续语音的声调评测算法,该算法可应用于计算机辅助语言学习系统和普通话水平测试中的声调评测。考虑到连续语音声调受上下文之间的相互影响,采用三音节单元建立高斯混合模型(Gaussian Mixture Model, GMM),三音节中辅音部分用Spline插值法拟合声调曲线来反映音节间基音频率的转移信息,并利用Fujisaki模型去除语句的语调和说话人个性特征,只对基频曲线中的声调特征建模。实验结果显示,相比于传统方法,采用三音节Spline插值和Fujisaki改进特征的方法使得机器与人工打分的相似度在测试集中分别提高了8.75%和14.09%。     

A comparative study on the performance of several speech recognition techniques applied on the highly confusing mandarin syllables

Abstract

In this paper, the performance of several speech recognition techniques applied on the highly confusing Mandarin syllables were carefully compared, including dynamic time warping (DTW), the newly proposed DTW with superimposed weighting function (DTWW), the discrete hidden Markov models (DHMM) and the continuous hidden Markov models (CHMM). The vocabulary used here consists of 409 first tone isolated Mandarin syllables. Due to the fact that many confusing sets exist in this vocabulary, the accurate recognition of these syllables is relatively difficult, and all the recognition experiments were performed in the speaker dependent mode. After a series of 13 experiments, it was found that the recognition rate of the newly proposed DTWW (88.3) is higher than that of DTW (85.1), DHMM (65.0) and CHMM (83.9), and that the CPU time used for DTWW is 1.03 times that for DTW, 24 times that for DHMM and 4.3 times that for CHMM. In addition, the memory space required for DTWW and DTW is 3.4 times that of DHMM and 8.5 times that of CHMM. Therefore, DTWW has the highest recognition rate, DHMM has the fastest recognition speed, whereas CHMM appears to be very attractive when all the different factors including recognition rate, recognition speed and memory space requirement are considered.