This paper introduces a combinational feature extraction approach to improve speech recognition systems. The main idea is to simultaneously benefit from some features obtained from Poincaré section applied to speech reconstructed phase space (RPS) and typical Mel frequency cepstral coefficients (MFCCs) which have a proved role in speech recognition field. With an appropriate dimension, the reconstructed phase space of speech signal is assured to be topologically equivalent to the dynamics of the speech production system, and could therefore include information that may be absent in linear analysis approaches. Moreover, complicated systems such as speech production system can present cyclic and oscillatory patterns and Poincaré sections could be used as an effective tool in analysis of such trajectories. In this research, a statistical modeling approach based on Gaussian mixture models (GMMs) is applied to Poincaré sections of speech RPS. A final pruned feature set is obtained by applying an efficient feature selection approach to the combination of the parameters of the GMM model and MFCC-based features. A hidden Markov model-based speech recognition system and TIMIT speech database are used to evaluate the performance of the proposed feature set by conducting isolated and continuous speech recognition experiments. By the proposed feature set, 5.7% absolute isolated phoneme recognition improvement is obtained against only MFCC-based features.
REFERENCES
1.
Abarbanel
, H.
, Analysis of Observed Chaotic Data
(Springer-Verlag
, New York
, 1996
).2.
Ashkenazy
, Y.
, Ivanov
, P. C.
, Havlin
, S.
, Peng
, C. K.
, Goldberger
, A. L.
, and Stanley
, H. E.
, “Magnitude and sign correlations in heartbeat fluctuations
,” Phys. Rev. Lett.
86
, 1900
–1903
(2001
).3.
Banbrook
, M.
, McLaughlin
, S.
, and Mann
, I.
, “Speech characterization and synthesis by nonlinear methods
,” IEEE Trans. Speech Audio Process.
7
, 1
–17
(1999
).4.
Bohez
, E.
and Senevirathne
, T. R.
, “Speech recognition using fractals
,” Pattern Recognition
(Elsevier
, 2001
), Vol. 34
, Iss. 11, pp. 2227
–2243
.5.
Brown
, R.
, “Calculating Lyapunov exponents for short and/or noisy data sets
,” Phys. Rev. E
47
, 3962
–3969
(1993
).6.
7.
Darbyshire
, A.
and Broomhead
, D.
, “Robust estimation of tangent maps and Lyapunov spectra
,” Physica D
89
, 287
–305
(1996
).8.
Dempster
, A. P.
, Laird
, N. M.
, and Rubin
, D. B.
, “Maximum likelihood from incomplete data via the EM algorithm
,” J. R. Stat. Soc. Ser. B (Methodol.)
39
, 1
–38
(1977
).10.
Garofolo
, J. S.
, Lamel
, L. F.
, Fisher
, W. M.
, Fiscus
, J. G.
, Pallett
, D. S.
, and Dahlgren
, N. L.
, DARPA TIMIT acoustic-phonetic continuous speech corpus [CD-ROM], 1993
.11.
Grassberger
, P.
and Procaccia
, I.
, “Measuring the strangeness of strange attractors
,” Physica D
9
, 189
–208
(1983
).12.
Hermansky
, H.
and Morgan
, N.
, “RASTA processing of speech
” IEEE Trans. Speech Audio Process.
2
, 578
–589
(1994
).13.
Johnson
, M. T.
, Povinelli
, R. J.
, Lindgren
, A. C.
, Ye
, J.
, Liu
, X.
, and Indrebo
, K. M.
, “Time-domain isolated phoneme classification using reconstructed phase spaces
,” IEEE Trans. Speech Audio Process.
13
, 458
–466
(2005
).14.
Kadtke
, J.
, “Classification of highly noisy signals using global dynamical models
,” Phys. Lett. A
203
, 196
–202
(1995
).15.
Kaiser
, J. F.
, “Some observations on vocal tract operation from a fluid flow point of view
,” in Vocal Fold Physiology: Biomechanics, Acoustics, and Phonatory Control
, edited by Titze
, I. R.
and Scherer
, R. C.
(Denver Center for Performing Arts
, Denver, CO
, 1983
), pp. 358
–386
.16.
Kantz
, H.
and Schreiber
, T.
, Nonlinear Time Series Analysis
(Cambridge University Press
, Cambridge, England
, 1997
).17.
Kokkinos
, I.
and Maragos
, P.
, “Nonlinear speech analysis using models for chaotic systems
,” IEEE Trans. Speech Audio Process.
13
, 1098
–1109
(2005
).18.
Kubin
, G.
, “Nonlinear processing of speech
,” in Speech Coding and Synthesis
, edited by Kleijn
, W. B.
and Paliwal
, K. K.
(Elsevier
, Amsterdam
, 1995
), Chap. 16.19.
Leung
, H.
, “System identification using chaos with application to equalization of a chaotic modulation system
,” IEEE Trans. Circuits Syst., I: Fundam. Theory Appl.
45
, 314
–320
(1998
).20.
Maragos
, P.
, Dimakis
, A.
, and Kokkinos
, I.
, “Some advances in nonlinear speech modeling using modulations, fractals, and chaos
,” Proceedings of the International Conference on DSP
, Santorini, Greece, July 2002
.21.
McGowan
, R. S.
, “An aeroacoustics approach to phonation
,” J. Acoust. Soc. Am.
83
, 696
–704
(1988
).22.
Mika
, S.
, Ratscht
, G.
, and Müller
, K. R.
, “A mathematical programming approach to the kernel Fisher algorithm
,” Advances in Neural Information Processing Systems
(MIT
, Cambridge, MA
, 2001
), Vol. 13
, pp. 591
–597
23.
Mika
, S.
, Ratscht
, G.
, Weston
, J.
, Scholkopft
, B.
, and Muller
, K. R.
, “Fisher discriminant analysis with kernels
,” IEEE Conference on Neural Networks for Signal Processing IX
, 1999
.47.
24.
25.
Munkherjee
, S.
, Osuna
, E.
, and Girosi
, F.
, “Nonlinear prediction of chaotic time series using support vector machines
,” Proceedings of the IEEE Workshop on Neural Networks for Signal Processing
, September 1997
, pp. 511
–520
.26.
Narayanan
, S.
and Alwan
, A.
, “A nonlinear dynamical systems analysis of fricative consonants
,” J. Acoust. Soc. Am.
97
, 2511
–2524
(1995
).27.
Ott
, E.
, Chaos in Dynamical Systems
(Cambridge University Press
, Cambridge, England
, 1993
).28.
Packard
, N. H.
, Crutchfield
, J. P.
, Farmer
, J. D.
, and Shaw
, R. S.
, “Geometry from a time series
,” Phys. Rev. Lett.
45
, 712
–716
(1980
).29.
Petry
, A.
, Augusto
, D.
, and Barone
, C.
, “Speaker identification using nonlinear dynamical features
,” Chaos, Solitons Fractals
13
, 221
–231
(2002
).30.
Pitsikalis
, V.
, Kokkinos
, I.
, and Maragos
, P.
, “Nonlinear analysis of speech signals: Generalized dimensions and Lyapunov exponents
,” Proceedings of Eurospeech
, Geneva, Switzerland, September 2003
.31.
Povinelli
, R. J.
, Johnson
, M. T.
, Lindgren
, A. C.
, Roberts
, F. M.
, and Ye
, J.
, “Statistical models of reconstructed phase spaces for signal classification
,” IEEE Trans. Signal Process.
54
, 2178
–2186
(2006
).32.
Povinelli
, R. J.
, Johnson
, M. T.
, Lindgren
, A. C.
, and Ye
, J.
, “Time series classification using Gaussian mixture models of reconstructed phase spaces
,” IEEE Trans. Knowl. Data Eng.
16
, 779
–783
(2004
).33.
Povinelli
, R. J.
, Roberts
, F. M.
, Ropella
, K. M.
, and Johnson
, M. T.
, “Are nonlinear ventricular arrhythmia characteristics lost, as signal duration decreases?
,” Proceedings of Computers in Cardiology
, Memphis, TN, 2002
, pp. 221
–224
.34.
Rapp
, P. E.
, Watanabe
, T. A. A.
, Faure
, P.
, and Cellucci
, C. J.
, “Nonlinear signal classification
,” Int. J. Bifurcation Chaos Appl. Sci. Eng.
12
, 1273
–1293
(2002
).35.
Rosenstein
, M. T.
, Collins
, J. J.
, and De Luca
, C.
, “A practical method for calculating largest Lyapunov exponents from small data sets
,” NeuroMuscular Research Center and Department of Biomedical Engineering, 1992
.36.
Sauer
, T.
, Yorke
, J. A.
, and Casdagli
, M.
, “Embedology
,” J. Stat. Phys.
65
, 579
–616
(1991
).37.
Sciamarella
, D.
and Mindlin
, G. B.
, “Unveiling the topological structure of chaotic flows from data
,” Phys. Rev. E
64
, 036209
(2001
).38.
Senevirathne
, T. R.
, Bohez
, E.
, and Van Winden
, J. A.
, “Amplitude scale method: New and efficient approach to measure the fractal dimension of speech wave forms
,” Electron. Lett.
28
, 420
–422
(1992
).39.
Skijarov
, O.
and Bortnik
, B.
, “Chaos and speech rhythm
,” Proceedings of IJCNN
, 2005
, Vol. 4
, pp. 2070
–2075
.40.
Takens
, F.
, “Detecting strange attractors in turbulence
,” Proceedings on Dynamical System Turbulence
, Warwick, 1980
, pp. 366
–381
.41.
Teager
, H. M.
and Teager
, S. M.
, “Evidence for nonlinear sound production mechanisms in the vocal tract
,” in Speech Production and Speech Modeling
, Vol. 55
, NATO Adv. Study Inst., Series D
, edited by Hardcastle
, W. J.
and Marchal
, A.
(Kluwer Academic
, Bonas, France
, 1990
), pp. 241–262.42.
Theiler
, J.
, Eubank
, S.
, Longtin
, A.
, Galdrikian
, B.
, and Farmer
, J. D.
, “Testing for nonlinearity in time series: The method of surrogate data
,” Physica D
58
, 77
–94
(1992
).43.
Tishby
, N.
, “A dynamical systems approach to speech processing
,” Proceedings of International Conference on Acoustics, Speech, Signal Processing (ICASSP)
, 1990
, pp. 365
–368
.44.
Vapnik
, V.
, Golowich
, S.
, and Smola
, A.
, “Support vector method for function approximation, regression estimation, and signal processing
,” Advances in Neural Information Processing Systems
(MIT
, Cambridge, MA
, 1996
), Vol. 8
, pp. 281
–287
.45.
Whitney
, H.
, “Differentiable manifolds
,” Ann. Math.
37
, 645
–680
(1936
).46.
Yamada
, M.
and Saiki
, M.
, “Chaotic properties of a fully developed model turbulence
,” Nonlinear Processes Geophys.
14
, 631
–640
(2007
).© 2010 American Institute of Physics.
2010
American Institute of Physics
You do not currently have access to this content.
