Matlab-based Formant Estimation

被引：2

作者：

Luo Jiao-yan ^{[1
]}

Sun Xiang-e ^{[1
]}

机构：

[1] Yangtze Univ, Sch Elect & Informat, Jinzhou, Hubei, Peoples R China

来源：

Applied Decisions in Area of Mechanical Engineering and Industrial Manufacturing | 2014年 / 577卷

关键词：

speech signal; formant; LPC; cepstrum;

D O I：

10.4028/www.scientific.net/AMM.577.798

中图分类号：

TH [机械、仪表工业];

学科分类号：

0802 ;

摘要：

In the study of speech digital signal processing, the resonance peak is the major characteristics in reflecting track resonant. In order to be more conducive to the future development of speech signal processing techniques, three methods of the formant estimation are introduced in the article, namely, Short-time Fourier Transform method, Cepstrum, LPC estimation method. These methods are compared and analyzed by means of MATLAB programming, and it is concluded that LPC method is the best method, which can estimate the formant peaks more accurate.

引用

页码：798 / 801

页数：4

共 9 条

[1]

Han jiqing, 2004, SPEECH SIGNAL PROCES, P12

[2]

Hu hang, 2009, SPEECH SIGNAL PROCES, P117

[3]

Huang Hai, 2006, Journal of Zhejiang University, V40, P1926

[4]

Lima Araujo A M, 1998, ITS, V1, P207

[5] Speech formant frequency estimation: evaluating a nonstationary analysis method [J].

Rao, P ;

Das Barman, A .

SIGNAL PROCESSING, 2000, 80 (08) :1655-1667

[6]

Wang Xiaoya, 2004, RADIOENGINEERING, V34, P59

[7]

Yang Shunliao, 2011, REALIZATION EXPT DIG

[8]

ZHAO QF, 1998, IEEE ISCAS, V5, P534

[9]

Zolfaghari P, 1996, ICSLP 96 - FOURTH INTERNATIONAL CONFERENCE ON SPOKEN LANGUAGE PROCESSING, PROCEEDINGS, VOLS 1-4, P1229, DOI 10.1109/ICSLP.1996.607830

← 1 →