Frequency and time filtering of filter-bank energies for HMM speech recognition

Author

Nadeu, Climenr ; Marino, José B. ; Hernando, Javier ; Nogueiras, Albino

Author_Institution

Univ. Politecnica de Catalunya, Barcelona, Spain

Volume

1

fYear

1996

fDate

3-6 Oct 1996

Firstpage

430

Abstract

In speech recognition, a discriminative frequency weighting can be achieved by decorrelating the frequency sequence of log mel-scaled filter-bank energies with a computationally inexpensive filter. We show how the spectral parameters that result from this kind of frequency filtering, both alone and combined with filtering of their time trajectories, are competitive with respect to the conventional cepstral representations of speech signals

Keywords

cepstral analysis; correlation methods; filtering theory; hidden Markov models; spectral analysis; speech recognition; HMM speech recognition; cepstral representations; computationally inexpensive filter; decorrelation; discriminative frequency weighting; frequency filtering; frequency sequence; hidden Markov model; log mel-scaled filter-bank energies; spectral parameters; speech signals; time filtering; Cepstral analysis; Cepstrum; Decorrelation; Discrete Fourier transforms; Filtering; Filters; Frequency domain analysis; Frequency estimation; Hidden Markov models; Speech recognition;

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.1996.607146

Filename

607146