3D vocal tract reconstruction using magnetic resonance imaging data to study fricative consonant production

Abstract

The development of Magnetic Resonance Imaging (MRI) has grown rapidly in clinical practice. Currently, the use of MRI in speech research provides useful and accurate qualitative and quantitative data of speech articulation. The aim of this work was to describe an effective method to extract vocal tract and compute their volumes during speech production from MRI images. Using a 3.0 Tesla MRI system, 2D and 3D images of the vocal tract were collected and used to analyze the vocal tract during the production of fricative consonants. These images were also used to build the associated 3D models and compute their volumes. This approach showed that, in general, the volumes measured for the voiceless consonants are smaller than the counterpart voiced consonants.