Experiments on a parametric nonlinear spectral warping for an HMM-based speech recognizer

Daniel J. Mashao

Experiments on a parametric nonlinear spectral warping for an HMM-based speech recognizer

Brown University

Research output: Contribution to journal › Conference article › peer-review

Abstract

This paper is concerned with the search for an optimal feature-set for a speech recognition system. A better acoustic feature analysis that suitably enhances the semantic information in a consistent fashion can reduce raw-score (no grammar) error rate significantly. A simple two-dimensional parameterized feature-set is proposed. The feature-set is compared against a standard mel-cepstrum, LPC-based feature-set in talker-independent, connected-alphadigit HMM-based recognizer. The results show that a particular combination of parameters yields a significantly lower error rate than the baseline mel-cepstrum LPC-based feature-set.

Original language	English
Pages (from-to)	17-20
Number of pages	4
Journal	Proceedings - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing
Volume	1
Publication status	Published - 1996
Externally published	Yes
Event	Proceedings of the 1996 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP. Part 1 (of 6) - Atlanta, GA, USA Duration: 7 May 1996 → 10 May 1996

ASJC Scopus subject areas

Software
Signal Processing
Electrical and Electronic Engineering

Cite this

@article{6554a330d1194401a440a526550755aa,

title = "Experiments on a parametric nonlinear spectral warping for an HMM-based speech recognizer",

abstract = "This paper is concerned with the search for an optimal feature-set for a speech recognition system. A better acoustic feature analysis that suitably enhances the semantic information in a consistent fashion can reduce raw-score (no grammar) error rate significantly. A simple two-dimensional parameterized feature-set is proposed. The feature-set is compared against a standard mel-cepstrum, LPC-based feature-set in talker-independent, connected-alphadigit HMM-based recognizer. The results show that a particular combination of parameters yields a significantly lower error rate than the baseline mel-cepstrum LPC-based feature-set.",

author = "Mashao, {Daniel J.}",

year = "1996",

language = "English",

volume = "1",

pages = "17--20",

journal = "Proceedings - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing",

issn = "0736-7791",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

note = "Proceedings of the 1996 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP. Part 1 (of 6) ; Conference date: 07-05-1996 Through 10-05-1996",

}

TY - JOUR

T1 - Experiments on a parametric nonlinear spectral warping for an HMM-based speech recognizer

AU - Mashao, Daniel J.

PY - 1996

Y1 - 1996

N2 - This paper is concerned with the search for an optimal feature-set for a speech recognition system. A better acoustic feature analysis that suitably enhances the semantic information in a consistent fashion can reduce raw-score (no grammar) error rate significantly. A simple two-dimensional parameterized feature-set is proposed. The feature-set is compared against a standard mel-cepstrum, LPC-based feature-set in talker-independent, connected-alphadigit HMM-based recognizer. The results show that a particular combination of parameters yields a significantly lower error rate than the baseline mel-cepstrum LPC-based feature-set.

AB - This paper is concerned with the search for an optimal feature-set for a speech recognition system. A better acoustic feature analysis that suitably enhances the semantic information in a consistent fashion can reduce raw-score (no grammar) error rate significantly. A simple two-dimensional parameterized feature-set is proposed. The feature-set is compared against a standard mel-cepstrum, LPC-based feature-set in talker-independent, connected-alphadigit HMM-based recognizer. The results show that a particular combination of parameters yields a significantly lower error rate than the baseline mel-cepstrum LPC-based feature-set.

UR - http://www.scopus.com/inward/record.url?scp=0029725521&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:0029725521

SN - 0736-7791

VL - 1

SP - 17

EP - 20

JO - Proceedings - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing

JF - Proceedings - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing

T2 - Proceedings of the 1996 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP. Part 1 (of 6)

Y2 - 7 May 1996 through 10 May 1996

ER -

Experiments on a parametric nonlinear spectral warping for an HMM-based speech recognizer

Abstract

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this