A Comparison of Reconstructed Phase Spaces and Cepstral Coefficients for Multi-Band Phoneme Classification

Authors

Kevin M. Indrebo, Marquette UniversityFollow
Richard J. Povinelli, Marquette UniversityFollow
Michael T. Johnson, Marquette UniversityFollow

Document Type

Conference Proceeding

Language

eng

Format of Original

4 p.

Publication Date

8-31-2004

Publisher

Institute of Electrical and Electronics Engineers (IEEE)

Source Publication

2004 7th International Conference on Signal Processing. Proceedings. ICSP '04

Source ISSN

0-7803-8406-7

Abstract

This paper examines the use of multi-band reconstructed phase spaces as models for phoneme classification. Sub-banding reconstructed phase spaces combines linear, frequency-based techniques with a nonlinear modeling approach to speech recognition. Experiments comparing the effects of filtering speech signals for both reconstructed phase space and traditional speech recognition approaches are presented. These experiments study the use of two non-overlapping subbands for isolated phoneme classification on the TIMIT corpus. It is shown that while classification accuracy using MeI frequency cepstral coefficients as features does not improve with sub-banding, the accuracy increases from 36.1% to 42.0% using sub-banded reconstructed phase spaces to model the phonemes.

Comments

Accepted version. Published as part of 2004 7th International Conference on Signal Processing. Proceedings. ICSP '04: 634-637. DOI. © 2004 The Institute of Electrical and Electronics Engineers. Used with permission.

Recommended Citation

Indrebo, Kevin M.; Povinelli, Richard J.; and Johnson, Michael T., "A Comparison of Reconstructed Phase Spaces and Cepstral Coefficients for Multi-Band Phoneme Classification" (2004). Electrical and Computer Engineering Faculty Research and Publications. 164.
https://epublications.marquette.edu/electric_fac/164