期刊文献+

检测语音端点及基音的概率模型及方法 被引量:4

Endpoint detection and pitch determination method based on a probability model
原文传递
导出
摘要 在传统的语音识别系统中,语音端点检测和基音提取是2个分开的步骤。将2个步骤合二为一将有助于提高误别准确性、降低系统复杂度。该文使用了一种语音时域分析方法,它使用概率模型描述语音幅度分布规律,并使用隐Markov模型(hidden Markov model,HMM)描述语音中的状态转换。使用新方法可以同时完成对语音端点的检测、清浊音的判断与基音频率计算。实验表明:这种算法在10dB以上信噪比的条件下可以得到准确的基音频率和端点位置。 Endpoint detection and pitch extraction are separated steps in traditional speech recognition systems.Combining these two steps together will improve precision and reduce complexity.A time domain analysis method is developed to describe the speech signal amplitude with a probability model and to model the state change of the speech with a hidden Markov model.The method simultaneously extracts the speech pitch and endpoints.Tests show that this algorithm can precisely detect the speech pitch frequency and endpoints for databases with a signal to noise ratio(SNR) more than 10 dB.
作者 胡波 肖熙
出处 《清华大学学报(自然科学版)》 EI CAS CSCD 北大核心 2013年第6期749-752,共4页 Journal of Tsinghua University(Science and Technology)
关键词 端点检测 基音提取 概率模型 统计方法 隐Markov模型(HMM) endpoint detection pitch extraction probability model statistics approach hidden Markov model(HMM)
作者简介 胡波(1988-),男(汉),甘肃。 通信作者:肖熙,副研究员,E-mail:xiaoxi@tsinghua.edu.cn
  • 相关文献

参考文献12

  • 1Li Q, Zheng J, Tsai A, et al. Robust endpoint detection and energy normalization for real-time speech and speaker recognition [J]. Speech and Audio Processing, 2002, 10(3) : 146 - 157.
  • 2Evangelopoulos G, Maragos P. Multiband modulation energy tracking for noisy speech detection [J]. Audio, Speech, and Language Processing, 2006, 14(6):2024 - 2038.
  • 3Shafran I, Rose R. Robust speech detection and segmentation for real-time ASR applications [C]// 2003 IEEE Conf Acoustics, Speech, and Signal Processing (ICASSP'03). Hong Kong, China, 2003.
  • 4罗世谦,冯子亮,张恒.一种基于能量聚类分析的句子语音端点检测法[J].计算机技术与发展,2008,18(4):13-15. 被引量:5
  • 5SHEN Jialin, Hung J, Lee L. Robust entropy-based endpoint detection for speech recognition in noisy environments [C]// ICSLP. Sydney, Australia, 1998.
  • 6Jia C, Xu B. An improved entropy based endpoint detection algorithm [C]// ISCSLP 2002. Taipei, China, 2002.
  • 7Wu B, Wang K. Robust endpoint detection algorithm based on the adaptive band-partitioning spectral entropy in adverse environments [J]. Speech and Audio Processing, 2005, 13(5): 762 - 775.
  • 8Khurram W, Weaver K, Salam F. A robust algorithm for detecting speech segments using an entropie contrast [C]//The 45th Midwest Symposium on Circuits and Systems (MWSCAS-2002). Tulsa, OK, USA: IEEE, 2002.
  • 9HUANG Liangsheng, YANG Chungho. A novel approach to robust speech endpoint detection in car environments [C]// 2000 IEEE Conf Acoustics, Speech, and Signal Processing (ICASSP'00). Istanbul, Turkey: IEEE, 2000.
  • 10徐大为,吴边,赵建伟,刘重庆.一种噪声环境下的实时语音端点检测算法[J].计算机工程与应用,2003,39(1):115-117. 被引量:30

二级参考文献8

  • 1Lee C H,Automatic Speech and speaker recognition-advanced topics,1996年
  • 2胡广书.数字信号处理[M].北京:清华大学出版社,2003..
  • 3古井贞熙.数字声音处理[M].朱家新,张国海,易武秀,译.北京:人民邮电出版社,1993.
  • 4Rabiner L, Juang Biing- Hwang. Fundamentals of Speech Recognition[ M]. [s.l.] :PTR Prentice- Hall, Inc, 1993.
  • 5吴亚栋.语音识别基础[D].上海:上海交通大学,1999.
  • 6张新字.Windows声音应用程序开发指南[M].西安:西安电子科技大学出版社,2003.
  • 7MHSavoji.ARobustAlgorithmforAccurateEndpointingofSpeechJ[].Space Communications.1989
  • 8李祖鹏,姚佩阳.一种语音段起止端点检测新方法[J].电讯技术,2000,40(3):68-70. 被引量:24

共引文献101

同被引文献25

  • 1Ghosh P K, Tsiartas A, Narayanan S. Robust Voice Activity Detection Using Long-Term Signal Variability [J]. IEEE Transactions on Audio, Speech and Language Processing, 2011, 19(3) : 601-613.
  • 2Cheng Gong, Zhang Xiongwei, Li Yaobo, et al. Voice Activity Detection Method Based on Gray Correlation Analysis Algorithm [J]. Journal of PLA University of Science and Technology, 2012, 56(7) : 1014-1022.
  • 3Cho N, Kim E K. Enhanced Voice Activity Detection Using Acoustic Event Detection and Classification [J]. IEEE Transactions on Consumer Electronics, 2011, 57(1) : 196-202.
  • 4Chiu Y H B, Raj B, Stern R M. Learning-Based Auditory Encoding for Robust Speech Recognition [J]. IEEE Transactions on Audio, Speech and Language Processing, 2012, 20(3) : 900-914.
  • 5Chun J, Sying J, Zhang R Trang. Tone recognition using ex- tended segments [J]. ACM Transactions on Asia Language In- formationprocessing, 2008, 7 (3): 1-23.
  • 6He Ba, Na Yang, Ilker Demirkol, et al. BaNa: A hybrid ap- proach for noise resilient pitch detection [C] //IEEE Statistical Signal Processing Workshop, 2012: 369-372.
  • 7Wang Ziwei, Yang Yingyun. Fast pitch detection based on im- proved FCM algorithm and hough transform [C]//Seventh In- ternational Conference on Image and Graphics, 2013.
  • 8He Jiao, He Zhimi, Xie Chaocheng. Pitch detection algo- rithm based on NCCF and CAMDF [C] //Fourth Internatio- nal Conference on Computational Intelligence and Communica- tion Networks, 2012.
  • 9陈小利,徐金甫.利用小波变换加权自相关的基音检测法[J].数据采集与处理,2007,22(4):463-467. 被引量:12
  • 10徐金甫,陈小利.基于线性预测的综合基音检测法[J].计算机工程与设计,2008,29(4):891-893. 被引量:6

引证文献4

二级引证文献35

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