Cepstrum-based pitch detection using a new statistical V/UV classification algorithm

Author

Ahmadi, Siavash ; Spanias, A.S.

Author_Institution

Nokia Mobile Phones Inc., San Diego, CA

Volume

7

Issue

3

fYear

1999

fDate

5/1/1999 12:00:00 AM

Firstpage

333

Lastpage

338

Abstract

An improved cepstrum-based voicing detection and pitch determination algorithm is presented. Voicing decisions are made using a multifeature voiced/unvoiced classification algorithm based on statistical analysis of cepstral peak, zero-crossing rate, and energy of short-time segments of the speech signal. Pitch frequency information is extracted by a modified cepstrum-based method and then carefully refined using pitch tracking, correction, and smoothing algorithms. Performance analysis on a large database indicates considerable improvement relative to the conventional cepstrum method. The proposed algorithm is also shown to be robust to additive noise

Keywords

cepstral analysis; feature extraction; frequency estimation; signal classification; signal detection; smoothing methods; speech processing; statistical analysis; additive noise; cepstral peak; cepstrum-based pitch detection; large database; modified cepstrum-based method; multifeature voiced/unvoiced classification algorithm; objective error measures; performance analysis; pitch determination algorithm; pitch frequency information; pitch tracking; short-time segments energy; smoothing algorithms; speech signal; statistical V/UV classification algorithm; statistical analysis; voicing detection; zero-crossing rate; Cepstral analysis; Cepstrum; Classification algorithms; Data mining; Databases; Frequency; Performance analysis; Smoothing methods; Speech analysis; Statistical analysis;

fLanguage

English

Journal_Title

Speech and Audio Processing, IEEE Transactions on

Publisher

ieee

ISSN

1063-6676

Type

jour

DOI

10.1109/89.759042

Filename

759042