期刊文献+

利用极点轨迹图探讨语速对语音共振峰的影响

A research concerning the effect of speaking rate on formant frequencies of speech using pole-locus plots
在线阅读 下载PDF
导出
摘要 基于语音共振峰频率与声道系统的极点存在一一对应的关系,针对语速变化导致语音参数变化的问题,提出了利用语音极点轨迹图探讨不同语速对共振峰影响的方法并进行了实验。实验中利用逆滤波器分别对快慢2种语速的单音节及连接数字语音提取极点并形成语音极点轨迹图。其快慢2种语速下的单音节语音的极点轨迹基本一致;对于数字连接词,比起快速语音,慢速语音的极点轨迹倾向于有更大的动态范围,即共振峰频率在发音过程中经历了更多变化。实验结果表明,对于孤立发音的单音节语音,语速变化对共振峰参数并无显著影响;而对于连接词语音,语速变化对共振峰参数有明显影响,慢速连接词语音的共振峰发生了更多变化。 Based on the corresponding relations between formant frequencies of speech and poles of vocal system,for the changes of speech signal parameters led by changes of speaking rate,a method of investigating the effect of speaking rate on formant frequencies is presented and related experiments are carried out. Using Inverse Filters,poles of both monosyllable and connected digits are extracted at fast and slow speaking rate respectively. For every syllable or connected digits,poles of all frames form a pole-locus plot. The results of experiments indicate that( 1) pole-locus plots of a monosyllabic speech with different speaking rates are rather similar and( 2) for connected digital speech,comparing with fast speech,pole-locus plots of slow speech have broader dynamic extent,namely formant frequencies changes much more during uttering. Thus the conclusion is obtained: speaking rate change has no significant effect on formant frequencies of isolated monosyllable speech but very noticeable effect on connected word speech which has much more changes of formant frequencies of slow speech.
出处 《北京信息科技大学学报(自然科学版)》 2015年第5期57-60,共4页 Journal of Beijing Information Science and Technology University
基金 北京市教委科技发展基金项目(KM200410772004)
关键词 语速 语音共振峰 语音参数 逆滤波器 极点轨迹图 speaking rate speech formant speech parameter inverse filter pole-locus plot
  • 相关文献

参考文献8

  • 1Joseph C Toscano,Bob Mc Murray.The timecourse of speaking rate compensation:effects of sentential rate and vowel length on voicing judgments[J].Language,Cognition and Neuroscience,2015,30(5):529-543.
  • 2许友亮,张连海,张文林,李永彬.基于语速调整和音位属性后验概率的音素识别[J].信号处理,2012,28(2):295-300. 被引量:5
  • 3Travis Wade,Bernd Mbius.Speaking rate effects in a landmark-based phonetic exemplar model[C].Proceedings of the 8th Annual Conference of the International Speech Communication Association(Interspeech2007).Antwerp,Belgium,2007:402-405.
  • 4孟君,杨大利.说话人辨认中通用背景模型训练时长研究[J].北京信息科技大学学报(自然科学版),2013,28(3):87-91. 被引量:4
  • 5Watanabe A.Formant estimation method using inverse-filter control[J].Speech and Audio Processing,IEEE Transactions on,2001,9(4).:317–326.
  • 6章文义,朱杰,陈斐利.一种新的共振峰参数提取算法及在语音识别中的应用[J].计算机工程,2003,29(13):67-68. 被引量:3
  • 7蔡莲红、黄德智、蔡锐.现代语音技术基础与技术[M].北京:清华大学出版社,2003:169-171.
  • 8Lawrence Rabiner,Biing-Hwang Juang.Fundamentals of speech recognition[M].影印本.北京:清华大学出版社,1999:28-29.

二级参考文献27

  • 1Chin-Hui Lee,Mark A.Clements,Sorin Dusan.An Overview on Automatic Speech Attribute Transcription(ASAT) [C]// Conference on the International Speech Communication Association.Antwerp,Belgium;InterSpeech Express, 2007.1825-1828.
  • 2S.King,P.Taylor.Detection of phonological features in continuous speech recognition using neural networks[J]. Computer,Speech and Language,2000,14(4):333-353.
  • 3M.A.Siegler,R.M.Stern.On the effects of speech rate in large vocabulary speech recognition systems[C]// International Conference on Acoustics,Speech,and Signal Processing. Detroit,MI:ICASSP express,1995.612-615.
  • 4V.R.Gadde,K.Sonmez,H.Franco.Multirate ASR Models for Phone-class Dependent N-best List Rescoring [C]//IEEE Workshop on Automatic Speech Recognition and Understanding(ASRU ).San Juan:IEEE express, 2005.157-161.
  • 5S.Dimopoulos,A.Potamianos,E.-F.Lussier,L.Chin-Hui. Multiple time resolution analysis of speech signal using MCE training with application to speech recognition [C]// International Conference on Acoustics,Speech, and Signal Processing.Tai Bei:IEEE express,2009. 3801-3804.
  • 6I-F Chen,Hsin-Min Wang.Articulatory Feature Asynchrony Analysis and Compensation in Detection-Based ASR//.International Speech Communication Association, Brighton United Kingdom,2009:3059-3062.
  • 7Zoltan Tuske,Christian Plahl,Ralf Schluter.A study on Speaker Normalized MLP Features in LVCSR[C]//Conference on the International Speech Communication Association. Florence,Italy,2011:1089-1092.
  • 8N.Strom,.“The NICO Artificial Neural Network Toolkit”, http://nico.nikkostrom.com.
  • 9Frantisek Grezl.Trap-Based Probabilistic Features For Automatic Speech Recognition[D].Brno,CZ:Brno University of Technology,2007.
  • 10Afsaneh Asaei,Benjamin Picart,Herve Bourlard.Analysis of Phone Posterior Feature space Exploiting Class-Specific Sparsity And MLP-Based Similarity Measure[C]// International Conference on ICASSP.Dallas,TX:2010. 4886-4889.

共引文献9

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部