InterSpeech 2021

A Simplified Model for the Vocal Tract of [s] with Inclined Incisors
(3 minutes introduction)

Tsukasa Yoshinaga (Toyohashi Tech, Japan), Kohei Tada (Toyohashi Tech, Japan), Kazunori Nozaki (Osaka University Dental Hospital, Japan), Akiyoshi Iida (Toyohashi Tech, Japan)
To examine the effects of inclined incisors on the phonation of [s], a simplified vocal tract model is proposed, and the acoustic characteristics with different maxillary incisor angles are predicted by the model. As a control model, a realistic vocal tract replica of [s] was constructed from medical images, and the angle of the maxillary incisor was changed from the original position up to 30°. The simplified model was constructed with a rectangular flow channel using the average dimensions of the vocal tracts for five Japanese subjects. Both geometries were set in an anechoic chamber, and sounds generated from the geometries were recorded with a microphone. The results showed that amplitudes of the sound generated by the realistic geometry were decreased by increasing the incisor angle, and this tendency agreed well with the simplified model. Moreover, the slope value of the decrease in overall pressure levels estimated by the model was consistent with that of the realistic geometry, indicating the capability of estimating the effects of inclined incisors with dental prostheses on the production of [s] by using the simplified model.