last updated: Feb. 9 2017
Speech science laboratory of Konan University
Vibration measurement on facial surfaces during speech production
Origins of speaker individualities
Speech sound is radiated not only from the mouth and nostril
openings but also from the surfaces of the head and neck. We
therefore measure vibration velocity pattern of facial surfaces
during phonation using laser-Doppler vibrometers. The vibration
pattern may be helpful as visual feedback of speaking and singing
Production and perception of non-linguistic information in speech
Everyone has unique voice characteristics as well as unique
face. We're investigating the origins of speaker individualities
in the speech production system based on magnetic resonance imaging
(MRI) technique. We're also developing a speech synthesis system that
can reproduce individual voice.
Acoustical characteristics of the vocal tract
Speech conveys not only linguistic information but also
non-linguistic one, speaker's sex, age, emotion, health condition,
and so forth. In this project, we explore how emotional speech is
produced, what acoustic features are used to identify speakers,
and how non-linguistic information is processed in our brain.
The transfer function of the vocal tract tells us about
relationship between vocal tract shape and acoustic features of
speech. So, we measure vocal tract shape during speech production
using MRI scanners, form realistic models of the vocal tract, and
measure their transfer function. In addition, we try to visualize
acoustic phenomena in the vocal tract by acoustic simulation.
Last update on Feb. 9 2017.
copyright © 2007-2017 Tatsuya Kitamura / All rights reserved.
ttyktmr [atmark] gmail.com