The use of wavelet transforms in phoneme recognition

Author

Tan, Beng T. ; Fu, Minyue ; Spray, Andrew ; Dermody, Phillip

Author_Institution

Dept. of Electr. & Comput. Eng., Newcastle Univ., NSW, Australia

Volume

4

fYear

1996

fDate

3-6 Oct 1996

Firstpage

2431

Abstract

The study investigates the usefulness of wavelet transforms in phoneme recognition. Both discrete wavelet transforms (DWT) and sampled continuous wavelet transforms (SCWT) are tested. The wavelet transform is used as a part of the front-end processor which extracts feature vectors for a speaker-independent HMM-based phoneme recognizer. The results are evaluated on a portion of the TIMIT corpus consisting of 30293 phoneme tokens for training and 14489 phoneme tokens for testing. The test results suggest that SCWT gives a considerably better recognition rate than DWT. On the other hand, the improvement of SCWT over Mel-scale cepstral coefficients appears to be marginal

Keywords

acoustic signal processing; feature extraction; hidden Markov models; speech processing; speech recognition; wavelet transforms; Mel-scale cepstral coefficients; TIMIT corpus; discrete wavelet transforms; feature vector extraction; front-end processor; phoneme recognition; phoneme tokens; recognition rate; sampled continuous wavelet transforms; speaker-independent HMM-based phoneme recognizer; testing; training; Cepstral analysis; Continuous wavelet transforms; Discrete wavelet transforms; Fourier transforms; Frequency; Sampling methods; Speech analysis; Testing; Wavelet analysis; Wavelet transforms;

fLanguage

English

Publisher

ieee

Conference_Titel

Spoken Language, 1996. ICSLP 96. Proceedings., Fourth International Conference on

Conference_Location

Philadelphia, PA

Print_ISBN

0-7803-3555-4

Type

conf

DOI

10.1109/ICSLP.1996.607300

Filename

607300