Flow-structure-acoustic interaction during a running speech

The production of a running speech is achieved by complex coordination of the laryngeal muscles which change the mechanical properties vocal folds to produce varied pitch and loudness through flow-structure-acoustics interactions. However, the dynamic coordination of laryngeal muscles during flow-structure-acoustic interaction is not fully studied. In this study, an MRI images based finite element (FE) vocal fold model is developed to perform the active muscle activations. The laryngeal structures, including all the muscles and cartilages, are constructed from MRI scan images. A previously proposed 1-D muscle activation model is integrated into 3-D FE code to model the contractile behaviors of the muscles. The model is further coupled with a flow-structure-acoustics interaction model to perform the simulation of running speeches. The activation of the muscles is simultaneously and independently controlled to produce simple English phrasal utterance, allowing investigation of interactions between the glottal flow, vocal folds and vocal tract in running speeches. The simulation model and the findings from this study would be helpful in fundamental research of voice, phonation modeling and surgical planning.