Study of Attractor Variation in the Reconstructed Phase Space of Speech Signals

Jinjin Ye; Michael T. Johnson; Richard J. Povinelli

Study of Attractor Variation in the Reconstructed Phase Space of Speech Signals

Jinjin Ye
, Michael T. Johnson
, Richard J. Povinelli

Research output: Contribution to conference › Paper › peer-review

2 Scopus citations

Abstract

This paper presents a study of the attractor variation in the reconstructed phase spaces of isolated phonemes. The approach is based on recent work in time-domain signal classification using dynamical signal models, whereby a statistical distribution model is obtained from the phase space and used for maximum likelihood classification. Two sets of experiments are presented in this paper. The first uses a variable time lag phase space to examine the effect of fundamental frequency on attractor patterns. The second focuses on speaker variability through an investigation of speaker-dependent phoneme classification across speaker sets of increasing size.

Original language	English
State	Published - 2003
Event	2003 ISCA Tutorial and Research Workshop on Nonlinear Speech Processing, NOLISP 2003 - Le Croisic, France Duration: May 20 2003 → May 23 2003

Conference

Conference	2003 ISCA Tutorial and Research Workshop on Nonlinear Speech Processing, NOLISP 2003
Country/Territory	France
City	Le Croisic
Period	5/20/03 → 5/23/03

Bibliographical note

Publisher Copyright:
© NOLISP 2003. All Rights Reserved.

Funding

This paper is based upon work supported by the National Science Foundation under Grant No. IIS-0113508.

Funders	Funder number
National Science Foundation (NSF)	IIS-0113508

ASJC Scopus subject areas

Communication
Artificial Intelligence
Signal Processing
Linguistics and Language

Cite this

@conference{5d859513cff344a78df1abc8cfbafe9a,

title = "Study of Attractor Variation in the Reconstructed Phase Space of Speech Signals",

abstract = "This paper presents a study of the attractor variation in the reconstructed phase spaces of isolated phonemes. The approach is based on recent work in time-domain signal classification using dynamical signal models, whereby a statistical distribution model is obtained from the phase space and used for maximum likelihood classification. Two sets of experiments are presented in this paper. The first uses a variable time lag phase space to examine the effect of fundamental frequency on attractor patterns. The second focuses on speaker variability through an investigation of speaker-dependent phoneme classification across speaker sets of increasing size.",

author = "Jinjin Ye and Johnson, \{Michael T.\} and Povinelli, \{Richard J.\}",

note = "Publisher Copyright: {\textcopyright} NOLISP 2003. All Rights Reserved.; 2003 ISCA Tutorial and Research Workshop on Nonlinear Speech Processing, NOLISP 2003 ; Conference date: 20-05-2003 Through 23-05-2003",

year = "2003",

language = "English",

}

TY - CONF

T1 - Study of Attractor Variation in the Reconstructed Phase Space of Speech Signals

AU - Ye, Jinjin

AU - Johnson, Michael T.

AU - Povinelli, Richard J.

PY - 2003

Y1 - 2003

N2 - This paper presents a study of the attractor variation in the reconstructed phase spaces of isolated phonemes. The approach is based on recent work in time-domain signal classification using dynamical signal models, whereby a statistical distribution model is obtained from the phase space and used for maximum likelihood classification. Two sets of experiments are presented in this paper. The first uses a variable time lag phase space to examine the effect of fundamental frequency on attractor patterns. The second focuses on speaker variability through an investigation of speaker-dependent phoneme classification across speaker sets of increasing size.

AB - This paper presents a study of the attractor variation in the reconstructed phase spaces of isolated phonemes. The approach is based on recent work in time-domain signal classification using dynamical signal models, whereby a statistical distribution model is obtained from the phase space and used for maximum likelihood classification. Two sets of experiments are presented in this paper. The first uses a variable time lag phase space to examine the effect of fundamental frequency on attractor patterns. The second focuses on speaker variability through an investigation of speaker-dependent phoneme classification across speaker sets of increasing size.

UR - https://www.scopus.com/pages/publications/22544465196

UR - https://www.scopus.com/pages/publications/22544465196#tab=citedBy

M3 - Paper

AN - SCOPUS:22544465196

T2 - 2003 ISCA Tutorial and Research Workshop on Nonlinear Speech Processing, NOLISP 2003

Y2 - 20 May 2003 through 23 May 2003

ER -

Study of Attractor Variation in the Reconstructed Phase Space of Speech Signals

Abstract

Conference

Bibliographical note

Funding

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this