Sigma-Lognormal Modeling of Speech.

Human movement studies and analyses have been fundamental in many scientific domains, ranging from neuroscience to education, pattern recognition to robotics, health care to sports, and beyond. Previous speech motor models were proposed to understand how speech movement is produced and how the resulting speech varies when some parameters are changed. However, the inverse approach, in which the muscular response parameters and the subject's age are derived from real continuous speech, is not possible with such models. Instead, in the handwriting field, the kinematic theory of rapid human movements and its associated Sigma-lognormal model have been applied successfully to obtain the muscular response parameters. This work presents a speech kinematics-based model that can be used to study, analyze, and reconstruct complex speech kinematics in a simplified manner. A method based on the kinematic theory of rapid human movements and its associated Sigma-lognormal model are applied to describe and to parameterize the asymptotic impulse response of the neuromuscular networks involved in speech as a response to a neuromotor command. The method used to carry out transformations from formants to a movement observation is also presented. Experiments carried out with the (English) VTR-TIMIT database and the (German) Saarbrucken Voice Database, including people of different ages, with and without laryngeal pathologies, corroborate the link between the extracted parameters and aging, on the one hand, and the proportion between the first and second formants required in applying the kinematic theory of rapid human movements, on the other. The results should drive innovative developments in the modeling and understanding of speech kinematics.

Biomechanical Description of Phonation in Children Affected by Williams Syndrome.

The voice of persons with Williams syndrome (WS) is described as hoarse with a deep and unstable fundamental frequency (f0). These observations may be justified by the deficit of elastin due to a haplo-insufficiency in the ELN gene characteristic of the syndrome. In view of the possible relationship between elastin deficit and dysphonia, a study of the dynamic function of WS phonation was conducted by means of biomechanical analysis. In order to assess the presence of dysphonic symptoms and their degree of severity, the biomechanical description of WS phonation has been evaluated in terms of dynamic mass and viscoelasticity estimates. Glottal biomechanical features such as vocal fold dynamic mass, stiffness, unbalances, and laryngeal tremor of 12 children with WS aged 3 to 8 years (five girls and seven boys) have been estimated and compared with the normative phonation of 97 children with typical development (53 girls and 44 boys). The results show that WS children show differences in f0, vocal fold mass and stiffness, phonation stability, glottal contact defects, and laryngeal tremor. The conclusions may help to make a more complete view of the connection between WS and dysphonia based on objective assessments.

Evaluation of voice pathology based on the estimation of vocal fold biomechanical parameters.

Voice disorders are a source of increasing concern as normal voice quality is a social demand for at least one third of the population in developed countries in cases where voice is an essential resource in professional exercise. In addition, the growing exposure to certain pathogenic factors such as smoking, alcohol abuse, air pollution, and acoustic contamination, and other problems such as gastro-esopharyngeal reflux or allergy as well as aging, aggravate voice disorders. Voice pathologies justify the assignment of larger resources to prevention policies, early detection, and less aggressive treatments. Traditional pathology detection relies on perceptive evaluation methods (GRABS), acoustic analysis, and visual inspection (indirect laryngoscopy, and modern fibro-endo-stroboscopy). This article describes a method for voice pathology detection based on the noninvasive estimation of vocal cord biomechanical parameters derived from voice using specific signal processing methods. Preliminary results using records from patients showing four frequent causes of voice pathology (nodules, polyps, chronic laryngitis, and Reinke's edema) are given. The results show that the alteration (distortion, unbalance, or deviation) of cord biomechanical parameters may serve as an indicator of pathology. Statistical methods based on hierarchical clustering and principal component analysis reveal that combining biomechanical estimates with classic perturbation parameters increases the accuracy of acoustic analysis, improving the detection of voice pathology. This research could open new possibilities for noninvasive screening of vocal fold pathologies and could be used in the implantation of e-health voice care services.

Automatic detection of voice impairments by means of short-term cepstral parameters and neural network based detectors.

It is well known that vocal and voice diseases do not necessarily cause perceptible changes in the acoustic voice signal. Acoustic analysis is a useful tool to diagnose voice diseases being a complementary technique to other methods based on direct observation of the vocal folds by laryngoscopy. Through the present paper two neural-network based classification approaches applied to the automatic detection of voice disorders will be studied. Structures studied are multilayer perceptron and learning vector quantization fed using short-term vectors calculated accordingly to the well-known Mel Frequency Coefficient cepstral parameterization. The paper shows that these architectures allow the detection of voice disorders--including glottic cancer--under highly reliable conditions. Within this context, the Learning Vector quantization methodology demonstrated to be more reliable than the multilayer perceptron architecture yielding 96% frame accuracy under similar working conditions.