Formant weighted cepstral feature for LSP-based speech recognition

Author

Hur, Ho Young ; Kim, Hyung Soon

Author_Institution

Dept. of Electron. Eng., Pusan Nat. Univ., South Korea

Volume

1

fYear

2001

fDate

2001

Firstpage

141

Abstract

In this paper, we propose a formant weighted cepstral feature for an LSP-based speech recognition system. The proposed weighting scheme is based on the well-known property of LSPs that the speech spectrum has a peak when adjacent LSFs come close. By applying this scheme to the pseudo-cepstrum (PCEP) conversion process (Kim et al. 1993), we can obtain formant weighted or peak enhanced cepstral features. Results of speech recognition experiments using QCELP coder output show that the proposed feature set outperforms the conventional features such as LSP or PCEP. Moreover its performance also exceeds that of the unquantized LPC cepstrum

Keywords

cepstral analysis; linear predictive coding; speech coding; speech recognition; LSP-based speech recognition; PCEP conversion process; QCELP coder output; formant weighted cepstral feature; line spectrum pair; liner predictive coding; peak enhanced cepstral feature; pseudo-cepstrum conversion process; speech spectrum; weighting scheme; Bit rate; Cepstral analysis; Cepstrum; Frequency; Linear predictive coding; Power harmonic filters; Speech coding; Speech processing; Speech recognition; Weight measurement;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 2001. Proceedings. (ICASSP '01). 2001 IEEE International Conference on

Conference_Location

Salt Lake City, UT

ISSN

1520-6149

Print_ISBN

0-7803-7041-4

Type

conf

DOI

10.1109/ICASSP.2001.940787

Filename

940787