Abstract:

Magnetic resonance imaging (MRI) has recently emerged as a useful tool for the investigation of the three-dimensional vocal tract shape in sustainable articulatory targets such as vowels and fricatives. However, existing scanning devices require that subjects phonate in a supine position. Previous work [Whalen, J. Acoust. Soc. Am. Suppl. 1 88 (1990)] compared MRI to cineradiographic data collected in a sitting posture, and reported limited effects of gravity on velar height differences in vowel production. This study used electromagnetometry [Perkell et al., J. Acoust. Soc. Am. 92, 3078--3096 (1992)] to investigate whether detectable systematic differences in midsagittal tongue posture exist between the two phonation conditions. Subjects fitted with nine transducers (four on the tongue, mandible, upper and lower lips, and two for head correction reference) produced both running speech and sustained vowels mimicking MRI protocol, repeated in sitting and supine postures. Analysis of separately recorded acoustics showed systematic effects of production posture on lower formant bandwidth. Results of comparing tongue height differences between the two conditions will be presented.