全部 标题 作者
关键词 摘要

OALib Journal期刊
ISSN: 2333-9721
费用:99美元

查看量下载量

相关文章

更多...

一种基于人耳听觉感知和子带补偿滤波的鲁棒语言辨识特征参数提取算法

, PP. 166-171

Keywords: 听觉感知,补偿滤波器,鲁棒性,语言辨识

Full-Text   Cite this paper   Add to My Lib

Abstract:

针对目前语言辨识系统所采用的特征参数没有充分考虑人耳听觉机制、鲁棒性较差的问题,提出一种符合人耳听觉感知特性的鲁棒语言辨识参数提取算法。该算法主要从两个方面提高特征参数的鲁棒性:在计算各子带能量时采用更符合人耳感知特性的Gammachirp滤波器组代替常用的三角滤波器组;为每一子带通道设计一个补偿滤波器。子带补偿滤波器的设计采用数据驱动的策略,通过补偿使得各子带滤波器输出信号的失真及环境噪音导致的失真同时达到最小。实验表明,文中所提出的特征在常见噪声环境下,性能均优于目前普遍使用的Mel频率倒谱系数特征及其衍生参数。关键词听觉感知,补偿滤波器,鲁棒性,语言辨识中图法分类号TN912。3ARobustFeatureParameterExtractionAlgorithmforLanguageIdentificationBasedonAudioPerceptionandSub-BandCompensationFilteringHUANGShan-Qi,ZHANGLing-Hai,QUDan(InstituteofInformationEngineering,InformationEngineeringUniversityofPLA,Zhengzhou450002)ABSTRACTIncurrentlanguageidentificationsystem,thecommonlyusedfeatureparametershavenotmadethebestuseofauditorycharacteristicsandhaveweakrobustnessincomplexenvironments。Anauditory-basedrobustfeatureextractionalgorithmisproposed。Eachsub-bandenergyoftheextractedauditoryfeaturesiscalculatedbyusingaGammachirpfilterbankinsteadofthecommonlyusedtrianglefilterbank。Thecompensationfilterusingdata-drivenanalysisforeachsub-bandoutputisobtainedbyaconstrainedoptimizationprocesswhichjointlyminimizestheenvironmentaldistortionaswellasthedistortioncausedbythefilteritself。ExperimentalresultsshowthatthefeatureoutperformstheMel-frequencycepstralcoefficientwidelyusedinnoisyenvironments。

References

[1]  Matejka P.Phonotactic and Acoustic Language Recognition.Ph.D Dissertation.Brno,Czech: Brno University of Technology,2008
[2]  Zhang Weiqiang,Liu Jia.Language Identification Based on Auditory Features.Journal of Tsinghua University: Science and Technology,2009,49(1): 78-81 (in Chinese) (张卫强,刘 加.基于听感知特征的语种识别.清华大学学报:自然科学版,2009,49(1): 78-81)
[3]  Wang Yue,Qian Zhihong,Wang Xue,et al.An Auditory Feature Extraction Algorithm Based on γ-Tone Filter-Banks.Acta Electronica Sinica,2010,38(3): 525-528(in Chinese)(王 玥,钱志鸿,王 雪,等.基于伽马通滤波器组的听觉特征提取算法研究.电子学报,2010,38(3): 525-528)
[4]  Kim C,Stern R M.Feature Extraction for Robust Speech Recognition Using a Power-Law Nonlinearity and Power-Bias Subtraction // Proc of the 10th Annual Conference of the International Speech Communication Association.Brighton,UK,2009: 28-31
[5]  Chiu Y H,Stern R M.Analysis of Physiologically-Motivated Signal Processing for Robust Speech Recognition // Proc of the 9th International Conference on Spoken Language.Brisbane,Australia,2006: 1000-1003
[6]  Chiu Y H B,Stern R M.Minimum Variance Modulation Filter for Robust Speech Recognition // Proc of the IEEE International Conference on Acoustics,Speech and Signal Processing.Taipei,China,2009: 3917-3920
[7]  Aertsen A M H J,Olders J H J,Johannesma P I M.Spectral-Temporal Receptive Fields of Auditory Neurons in the Grassfrog.Biological Cybernetics,1981,39(3): 195-209
[8]  Yan Luo.The Study of Filter Bank to Simulate the Characteristics of the Human Basilar Membrane.Master Dissertation.Beijing,China: Beijing Jiaotong University,2009 (in Chinese)(颜 罗.人耳基底膜滤波器仿真研究.硕士学位论文.北京:北京交通大学,2009)
[9]  Irino T,Patterson R D.A Time-Domain Level-Dependent Auditory Filter: The Gammachirp.The Journal of the Acoustical Society of America,1997,101(1): 412-419
[10]  Hajaiej Z,Ouni K,Ellouze N.Gammachirp Filter Frond-End for Automatic Speech Recognition // Proc of the 4th International Conference on Sciences of Electronic,Technologies of Information and Telecommunications.Hammamet,Tunisia,2007: 873-877
[11]  Torres-carrasquillo P A,Singer E,Gleason T,et al.The MITLL NIST LRE 2009 Language Recognition System // Proc of the IEEE International Conference on Acoustics Speech and Signal Processing.Dallas,USA,2010: 4994-4997

Full-Text

Contact Us

service@oalib.com

QQ:3279437679

WhatsApp +8615387084133