The palatalized articulation for the dental and alveolar sounds has been reported as a type of articulation errors in the cleft palate speech. In the present paper, articulatory tongue gestures for this particular type of error were observed by the use of dynamic palatograph and cineradiograph. The subjects were 9 postoperative cleft palate patients with this type of articulation error, 5 males and 4 females, ranging from 5 to 23 years old. Among the 9 subjects, 2 were observed by cineradiograph and 8 by dynamic palatograph. The result revealed the tongue gestures were characterized by the articulatory contact of the back of the tongue, instead of the tip or the blade, to the posterior end of the hard palate. The tongue-palate contact was clearly different from that for the normally palatalized /s/ which is represented as [∫] in Japanese. Sound spectrograms of the fricative sound produced by this abnormally palatalized articulation showed a marked decrease in the noise energy in the high frequency range as compared to the [s] or [∫] . The results supported the previous report where characteristic features of this articulation error were described mainly by auditory analysis and by limited visual observations from the mouth opening. Articulation training was administered to the patients in order to replace the faulty articulation by the correct gestures for the dental and alveolar sounds. The tongue-palate contact pattern which was nearly identical to that in normal subjects was observed when the patient reached the clinically acceptable goal.
The purpose of this paper is to discuss the possibility of application of ultrasonic technique to observe tongue shape. One of the advantages of ultrasonic technique is harmlessness to the human subject. Beside this, the frontal section of the tongue which is the difficult object for ordinary X-ray photo can be visualized. Sagittal contours of the tongue for five Japanese vowels observed by this method are well corresponding to those by X-ray technique. Contours of the frontal section at around the tongue root show a groove formation during /i/, /e/, and /o/ production. For all of these three vowels, the genioglossus muscle shows higher activity. Therefore, the observed groove may be representing the inserting portion of the genioglossus muscle. Since the apparatus we used in this study is designed and developed for the clinical purpose, for example to examine the thyroid tumor, the data we got had limited informations especially, the visualized area was narrow. But if we can use the sector type porbe instead of the linear type, we might have the wider area to be observed. The ultrasonic apparatus would be appropreate for observation of the tongue shape.