Second International Workshop on Models and Analysis of Vocal Emissions for Biomedical Applications (MAVEBA 2001)

Florence, Italy
September 13-15, 2001

A Physically-Informed Model of the Glottis with Application to Voice Quality Assessment

Carlo Drioli, Federico Avanzini

Dip. di Elettronica e Informatica, Università di Padova, Italy

A glottal model based on physical insights is proposed. The model describes the vocal fold as a second order oscillator, i.e. a damped mass-spring system. The oscillator is coupled with a non-linear block, accounting for fold interaction with airflow. The non-linear block is modeled as a regressor-based functional with weights to be identified, and a pitch-synchronous identification procedure is outlined. The model is used to analyze voiced sounds from normal and from pathological voices, and the application of the proposed method to voice quality assessment is discussed.

Index Terms. Non-linear models of vocal emission; physical models; estimation and modeling of the glottal flow; non-linear identification; voice quality assessment

Full Paper

Bibliographic reference.  Drioli, Carlo / Avanzini, Federico (2001): "A physically-informed model of the glottis with application to voice quality assessment", In MAVEBA-2001, 117-122.