Chaos in Voice, From Modeling to Measurement

The glottal to noise excitation ratio (GNE) is an acoustic measure designed to assess the amount of noise in a pulse train generated by the oscillation of the vocal folds. So far its properties have only been studied for synthesized signals, where it was found to be independent of variations of fundamental frequency (jitter) and amplitude (shimmer). On the other hand, other features designed for the same purpose like NNE (normalized noise energy) or CHNR (cepstrum based harmonics-to-noise ratio) did not show this independence. This advantage of the GNE over NNE and CHNR, as well as its general applicability in voice quality assessment, is now tested for real speech using a large group of pathologic voices (n=447). A set of four acoustic features is extracted from a total of 22 mostly well-known acoustic voice quality measures by correlation analysis, mutual information analysis, and principal components analysis. Three of these measures are chosen to assess primarily different aspects of signal aperiodici...

Selection and combination of acoustic features for the description of pathologic voices

Acoustic analyses of sustained and running voices from patients with laryngeal pathologies.

The chaos mechanism of above-range phonation was examined in symmetrically modeled vocal folds by using the traditional two-mass model assumption. The Poincare map technique was used to display chaotic attractors. This method provided an effective description of irregular vocal-fold movements. The power spectrum, Lyapunov exponent, and Kaplan–Yorke dimension were employed to describe chaotic vibrations in the vocal-fold model. These nonlinear dynamic analyses suggested that, for the positive Lyapunov exponent, chaotic attractors contribute to irregular vocal-fold vibrations. Descriptions of complicated irregular vibrations of the vocal fold yielded evidence of chaos. To investigate the effects of independent parameters such as subglottal pressure, coupling stiffness, and phonation neutral area, bifurcation diagrams based on the Poincare map were discussed. The results confirmed that the dynamics of the two-mass model was strongly influenced by independent parameters. Nonlinear dynamic methods were expected to provide useful information for better understanding of irregular vocal-fold vibrations as well as of the dynamic mechanism of above-range phonation in excised larynx experimentation.

Modeling of chaotic vibrations in symmetric vocal folds.

Phonatory impairment in Parkinson's disease: evidence from nonlinear dynamic analysis and perturbation analysis.

This paper reports results of the estimation of dynamical invariants, namely Lyapunov exponents, dimension, and metric entropy for speech signals. Two optimality criteria from dynamical systems literature, namely singular value decomposition method and the redundancy method, are used to reconstruct state space trajectories of speech and make observations. The positive values of the largest Lyapunov exponent of speech signals in the form of phoneme articulations show the average exponential divergence of nearby trajectories in the reconstructed state space. The dimension of a time series is a measure of its complexity and gives bounds on the number of state space variables needed to model it. It is found that most speech signals in the form of phoneme articulations are low dimensional. For comparison, a statistical model of a speech time series is also used to estimate the correlation dimension. The second‐order dynamical entropy (which is a lower bound of metric entropy) of speech time series is found to ...

Nonlinear dynamical analysis of speech

The objective of the present paper is the study of models which view the given 1-D data as the impulse response of a minimal order discrete linear filter with data and coefficients confined to a finite field. This representation is known as “minimal partial realization” problem in linear system theory and has been tackled successfully by Ho and Kalman. Interestingly, the same problem is addressed and solved in a coding theory context by the well-known Berlekamp-Massey (bm) algorithm. We present a modified recursive version of thebm algorithm specially suited to our task of signal modelling. By introducing an absolute error bound, one can further solve the approximation problem computationally. The paper includes a number of examples and a mention of the software for this purpose.

/pdf/modified-berlekamp-massey-algorithm-and-modelling-of-19o2g1ig9f.pdf

S. K. Mullick

Papers

Nonlinear dynamical analysis of speech

Modified Berlekamp-Massey algorithm and modelling of discrete-time signals