Phondi: Jon Yip on ultrasound, lip cameras and Greek consonant clusters


By Andries W. Coetzee
Oct 13, 2012 Bookmark and Share

Ultrasound image of the tongue during speech

Ultrasound image of the tongue during speech

Over the past two years we have added several new pieces of equipment to the Phonetics Laboratory that enables us to now also do detailed articulatory phonetic research. We have added an ultrasound machine that can be used to study tongue movement during speech. More recently, we also acquired a camera that can be used to monitor lip movement during speech. Our graduate students are taking the lead in figuring out how all of these new technologies work and how to use them to conduct phonetic research.

Jon Yip, for instance, is using both the lip camera and the ultrasound machine in his dissertation research on the gestural timing patterns in consonant clusters in Modern Greek. This Friday in Phondi (10/19, 1 pm, Lorch 473), Jon will present some of the results of his dissertation research - as a practice run for a presentation that he will give later in October at the 164th Meeting of the Acoustical Society of America in Kansas City. If you are interested in articulation or just want to learn about the new additions to the Phonetics Laboratory and how they are helping us to explore new topics, come to Phondi this Friday.

Linguistic effects on the timing of gestural coordination in Modern Greek CC sequences

While position in the word, order of place of articulation, and manner of articulation have been shown in various languages to influence gestural timing between consonant sounds, at present it is unclear whether these effects are caused by biomechanical or perceptual-recoverability constraints on production.  This study, which focuses on Modern Greek speakers’ CC productions, tests whether the source of the effects of within-word position (initial, medial), place order (front-to-back, back-to-front), C1 manner (plosive, fricative), and C2 manner (plosive, fricative, liquid) on articulatory lag between C1 and C2 are biomechanical or perceptual in nature.  A perception-oriented account predicts an influence of both C1 and C2 manner on gestural lag when acoustic masking is most likely (when a CC is word-initial and back-to-front), whereas a biomechanical account predicts no such effect.  In order to assess relative degrees of gestural lag, ultrasound imaging and lip video are used to track the timing patterns of labial, tongue-tip, and tongue-dorsum constrictions simultaneously during production.  Supporting both hypotheses, a global effect of place order was found, that is, gestural lag is greater in back-to-front [kt ks kl xt] than in front-to-back [pt ps pl ft].  However, gestural lag measures across C1+C2 manners (i.e. plosive-plosive, plosive-fricative, plosive-lateral, fricative-plosive) show a pattern that is dependent on place order.  Namely, C1-C2 manner does not matter unless the place order is front-to-back, not the other way around as predicted under the perceptual-recoverability account.