Excitation modeling in a homomorphic vocoder

Author

Chung, Jae H. ; Schafer, Ronald W.

Author_Institution

Sch. of Electr. Eng., Georgia Inst. of Technol., Atlanta, GA, USA

fYear

1990

fDate

3-6 Apr 1990

Firstpage

25

Abstract

An efficient dynamic excitation modeling which emphasizes either the pitch-dependent periodic nature or the more random nature of the sound sources depending on the voicing state of the speech signal is applied to the homorphic vocoder. In this excitation modeling scheme, the cepstral peak value is used as the basis for classifying a given speech segment into three different classes: voiced, unvoiced, and mixed. Different excitation modeling strategies are used for the different classes. The algorithm is simulated, and clear improvement in the speech quality is shown

Keywords

encoding; speech analysis and processing; vocoders; cepstral peak value; excitation modeling scheme; excitation modeling strategies; homomorphic vocoder; low bit rate vocoders; mixed voiced speech; pitch-dependent periodic nature; random nature; speech encoding; speech quality; speech segment classification; three classes; unvoiced speech; voiced speech; voicing state; Analytical models; Bit rate; Cepstral analysis; Cepstrum; Deconvolution; Filters; Speech; Speech analysis; Speech coding; Speech synthesis; Vocoders;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 1990. ICASSP-90., 1990 International Conference on

Conference_Location

Albuquerque, NM

ISSN

1520-6149

Type

conf

DOI

10.1109/ICASSP.1990.115528

Filename

115528