基于跟踪共振峰的语音增强算法

杨凌; 杨海波; 高新春

doi:10.3724/SP.J.1146.2008.01361

基于跟踪共振峰的语音增强算法

doi: 10.3724/SP.J.1146.2008.01361

基金项目:

甘肃省自然科学基金(0803RJZA026)资助课题

计量
- 文章访问数: 3490
- HTML全文浏览量: 99
- PDF下载量: 1051
- 被引次数: 0
出版历程
- 收稿日期: 2008-10-17
- 修回日期: 2009-04-09
- 刊出日期: 2009-10-19

Speech Enhancement Algorithm Based on Tracing the Formant

摘要

摘要: 该文通过实验方法研究和分析了汉字语音共振峰的特点，发现可跟踪并找到各个共振峰，结合汉字发音所具有的一般规律，提出了一种基于跟踪共振峰的语音增强算法。该算法能够有效地识别出带噪语音中的语音帧和非语音帧，简单且有效地去除非语音帧的全部噪声，明显抑制语音帧内的噪声。算法计算复杂度低并具有噪声环境可移植性。
- 语音信号处理; 共振峰; 共振峰区; 语音帧
Abstract: It is obtained that every formant can be traced and found out by studying and analyzing the Chinese speech in the way of experiment in this paper, combining with the generally articulation feature of the Chinese speech, a speech enhancement algorithm based on tracing the formant is proposed. The speech frames can be determined from the non-speech frames effectively using the algorithm, and then, all the noise in the non-speech frames can be wiped off simply and validly, and noise in the speech frame can be degraded largely. Also, the algorithm has some other good characteristics, such as simple computation and noise-environment transplantability and so on.

HTML全文

参考文献(1)

Boll S F. Suppression of acoustic noise in speech usingspectral subtraction[J].IEEE Transactions AcousticsSpeech, Signal Processing.1979, 27(4):113-120[2]Berouti M, Schwartz R, and Makhoul J. Enhancement ofspeech corrupted by acoustic noise[C]. Proc. of IEEE ICASSP,Washington, DC, USA, Apr 1979: 208-211.[3]Lim J S and Oppenheim A V. Enhancement and bandwidthcompression of noisy speech[J].Proc. of the IEEE.1979,67(12):1586-1604[4]McAulay R J and Malpass M L. Speech enhancement using asoft-decision noise suppression filter[J].IEEE TransactionsAcoustics, Speech, Signal Processing.1980, 28(4):137-145[5]Ephraim Y and Malah D. Speech enhancement using aminimum-mean square error short-time spectral amplitudeestimator[M][J].IEEE Transactions on Acoustics. Speech SignalProcessing.1984, 32(12):1109-1121[6]Wang Guang-yan, Wang Xia, and Zhao Xiao-qun. Speechenhancement based on a combined spectral subtraction withspectral estimation in various noise environment[C]. IEEEInternational Conference on Audio, Language and ImageProcessing, ICALIP-2008, Shanghai, China, July 2008:1424-1429.[7]Diethorn E J. Subband Noise Reduction Methods for SpeechEnhancement[M]. Boston:Kluwer Academic Publishers, MA,2004: 91-115.[8]Ephraim Y and Van Trees H L. A signal subspace approachfor speech enhancement[J].IEEE Transactions on SpeechAudio Processing.1995, 3(7):251-266[9]Benesty J, Makina S, and Chenk J. Speech Enhancement[M].Berlin: Springer-Verlag, 2005: 1-8, 135-159.[10]张丽艳, 殷福亮. 一种改进的奇异值分解语音增强方法[J].电子与信息学报.2008, 30(2):357-361浏览[11]Paliwal K K and Basu A. A speech enhancement methodbased on Kalman filtering[C]. Proc. IEEE ICASSP, 1987:17-180.[12]Li Hui, Wang Xin, Dai Bei-qian, and Lu Wei. A kalmansmoothing algorithm for speech enhancement based on theproperties of vocal tract varying slowly[C]. IEEE EighthACIS International Conference on Software Engineering,Artificial Intelligence, networking, and Parallel/ DistributedComputing, SNPD-2007, Qingdao, China, July 2007, Vol.3:832-836.[13]You Chang-huai, Rahardja S, and Koh Soo Ngee.Autoregressive parameter estimation for Kalman filteringspeech enhancement[C]. IEEE International conference onAcoustics,Speech and Signal Processing, ICASSP-2007,Honolulu, Hl, America, April 2007, Vol.4: IV-913-IV-916.[14]Ruiz I, Garca B, Mndez A, and Villanueva V. Oesophagealspeech enhancement using Kalman filters[C]. IEEEInternational symposium on Signal Processing andInformation Technology, Giza, Egypt, Dec. 2007: 1176-1179.[15]Donoho D L. De-noising by soft thresholding[J].IEEETransactions on Information Theory.1995, 41(3):613-627[16]Shao Yu and Chang Chip-hong. A versatile speechenhancement system based on perceptual waveletdenoising[C]. IEEE International Symposium on Circuits andSystems, ISCAS-2005, Kobe, Japan, May 2005, Vol.2:864-867.[17]Xu Yao-hua, Wang Gang, Gu Ying, and Liu Hai-yang. Anovel wavelet packet speech enhancement algorithm based ontime-frequency threshold[C]. IEEE Second InternationalConference on Innovative Computing, Information andControl, ICICIC-2007, Kumamoto, Japan, September 2007:492.[18]Zhang L H and Rong G F. A kind of modified speechenhancement algorithm based on wavelet packagetransformation[C]. IEEE International Conference onWavelet Analysis and Pattern Recognition, ICWAPR-2008,Hong Kong, China, August 2008, Vol.1: 421-425.[19]Hsung Tai-chui and Lun Pak-Kong D. Speech enhancementbased on adaptive wavelet denoising on multitaperspectrum[C]. In IEEE International Symposium on Circuitsand Systems, ISCAS-2008, Seattle, America, May 2008:1700-1703.[20]徐耀华, 王刚, 郭英. 基于时频阈值的小波包语音增强算法[J].电子与信息学报.2008, 30(6):1363-1366浏览[21]杨行峻, 迟惠生. 语音信号数字处理[M]. 北京: 清华大学出版社, 1995: 13-20.[22]Hangsen J and Pellom B. An effective quality evaluationprotocol for speech enhancement algorithms[C]. Int. Conf.Spoken Language Processing, Sydney, Australia, Dec, 1998:2819-2922.

施引文献

资源附件(0)

访问统计