RPDOC - Retour Post-Doctorants

Vocal Folds Characterization and Modelling – VOFOCAM

Vocal folds Characterization and Modelling

The vocal folds are the main actors of the production of speech. They are heteregeneous structures, mainly controlled by the ligaments tension and the muscular activity, that can be degraded due to vocal effort or aging, potentially leading to voice disorders.


Qualitative as well as quantitative understanding of the physics of voice relies on the knowledge of the vibratory state of the vocal folds. Experimental set-ups have been designed to provide an insight on the phenomenon involved in the phonation in a controlled environment. Reproducibility and repeatability are key points in the research, requiring tools for a better observation of the behaviour of the involved structures along with the variation of a wide range of control parameters linked to physiological aspects.

Mechanical characterization of the materials and structures used in the set-ups are performed. From the experimentation point of view, it relies on the innovative combination of recent measurement techniques to fully acquire the motion of the membrane defining the glottis: laser doppler vibrometry and high-speed imaging are used to get both the axial and transverse components of the displacement of the membrane in the vicinity of the constriction, either externally driven or during the self-sustained oscillations. These results will be compared with those obtained by numerical simulation (finite element methods) for different thicknesses and assemblies of the membrane layers, tension and internal pressure of the vocal folds. The ability of numerically predict the behavior lead to simplified modeling of structures based on the number of degrees of freedom necessary to take into account a number of asymmetries of the vocal folds and the environment. This model ultimately provide a parametric space for the study of pathological configurations phonation.

First results showed the capacity of the model to reproduce pathological behavior by unilateral presence of cyst (hard or soft), resulting in a high onset oscillation (significant increase in pressure thresholds). On the other hand, plate models were used to simulate, using FE methods, the inflation of the layers of latex (as used in the set-up) due to internal pressure. However, the modal analysis leads to unconvincing results that led us to review and modify some aspects of modelling.

As mentionned in the scientific document, it is expected to acquire a vibrometry measurement device (requiring to adjust the set-up accordingly) to better understand the nature of the motion of the vocal folds and bring improvement to the socalled complete models, before attempting to isolate sufficient degrees of freedom able to represent the various asymmetries in the configuration and produce a minimum model for the quasi-analytical study of pathologies.

The work performed during the initial semester provided materials for an international communication (ICA'13 in Montréal Canada), and also a master's thesis, supported by ANR funding.

Voice production is the result of a complex interaction between the air expelled from the lungs, the vocal tract from the larynx to the lips and the vocal folds delimiting the glottis. The pulsatile flow generated during the self-sustained oscillations of the glottis wall tissues acts as an source for acoustic waves that will carry part of the speech information. The vocal folds behaves as a valve that periodically opens and closes the glottis, both modulating the airflow and sustained by the fluid-structure energy transfer. The existence of these oscillations is however only possible due to the jet formed at the glottal exit and by suitable vocal folds characteristics. These latter are constituted by several tissues layers that are controlled, consciously or not, by the ligament tension and the muscular activity, and subject to the action of the pressure force from the air flowing through the glottis. This complex heterogeneous structure may be degraded, leading to potential source of voice disorders.

The objective of the VOFOCAM project is to develop the tools that make possible to observe and analyse the structural health of the laryngeal tissues (vocal folds and ventricular bands). Vocal folds mechanical replicas as the one of the GIPSA-lab (UMR 5216, CNRS – Univ. Grenoble) have been set up to examine under controlled conditions the peculiarities of the glottal flow and its interaction with mechanical deformable structures and with the acoustic resonant vocal tract. The replicas provide a support for the development of an measurement device to capture the large deformation of the artificial vocal folds. Using a wide range of material, tension and positioning, we aim to built a large set of configurations that would be as representative as possible of the laryngeal pathologies that will analysed with mathematical methodologies accounting for the spatially distributed information. The reduced order model that could in particular represent anterior-posterior, left-right and subglottal-supraglottal asymmetries will induce the parametrization of the vibration patterns observed on the oscillating replica, and allow to build a database suitable for sound-synthesis purposes and that could be the basis of a human vocal folds analysis device in conjunction with the transposition of the developed experimental techniques in a clinical context.

Project coordination

Fabrice SILVA (Centre National de la Recherche Scientifique) – silva@crans.org

The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.


CNRS DR11 Centre National de la Recherche Scientifique

Help of the ANR 493,698 euros
Beginning and duration of the scientific project: October 2012 - 36 Months

Useful links

Explorez notre base de projets financés



ANR makes available its datasets on funded projects, click here to find more.

Sign up for the latest news:
Subscribe to our newsletter