885095976

can bc generalized to another place of articulation. Results to be pre-sented are of tongue movemcnt during closing and opening gestures of thc vclar stops in the following words: A/: locale lócal fóllicle; /g/: cigar vigor vinegar. [Work supported by Sigma Xi.) 2SP12. Toward the development of articulatory signatures for łntelligibility test words: Lingua! kinematics. Gary Weismer, Ray D. Kent, and Greg Turner (Dept. of Commun. Disord. and Waisman Ctr., Univ. of Wisconsin-Madison, 1500 Highland Ave., Madison, W! 53705-2280)

The devclopment of “normal" articulatory signatures for intelligi-bility test words should eventually permit highly specific designatioas of the articulatory basis of intelligibility deficits in persons with motor speech disorders. Our study of lingual kinematics for selected words, using x-ray microbeam data, suggests that a major obstacle to the con-struction of such signatures Ls the substantial interspeaker variabi)ity in pellet trajectories for a given word. This variability will be reported, ways in which it might bc minimized will be discussed, and some anal-yses of rebtionships between lingual kinematics and vocalic formant trajectories will be described. One conclusion that has emerged from analyses to datę is that typically used kinematie measurands such as peak Yelodty, instantaneous acceleration, and so forth, will not be very useful in the development of an articulatory signature. Rather, thc un-folding of articulatory gestures over time must somehow be incorpo-rated into the signature concept. [Work supported by NIH.) 2SP13. Linguo-raandibular coordination in consonant production.

Alice Faber (Haskins Labs., 270 Crown St., New Haven, CT 06511) and Edda Fametani (Centro di Fonetica del CNR, 35122 Padova, Italy)

The coordination between the jaw and tongue in regulating tongue position for consonant articulation was explored by simultaneous use of electropalatography and an altcmating magnetic field device (Movetrack: Branderud, 1985). Linguo-palatal constriction was moni-tored via palatography, and tongue body and jaw hcight and frontness were monitored by smali receiver coils adhering to the arliculators. Speech materials were real and nonsense words of the form (C)V,CV₂, where C ranged ovcr {t d / z 1} and V, and V₂ over {a i}. The “real” words were produced in isolation and in sentences. Gach token was repeated three times by one femalc native speaker of Italian. Regression analysis was used both to distinguish between rotational and transla-tional movement of the jaw and to sępa ratę between the tongue's independent movement and positional effects ca used by jaw movement. Ex-amination of the inherent differences among consonants in tongue position reveals both synergistic and antagonistic pat tern s of linguo-mandibular coordination. The tongue body is lower for /z/ than for the other consonants, and fronted for (in order) /z JY> /t/> /d I/. Com-parison with the jaw positions and with the residuals from the regression analyses reveals that the tongue height differences result from a combi-nation of jaw height differences (/z J/>/t d/>/l/) and active lower-ing of the tongue body (relative to the jaw) for /z J/. Similarly, the tongue frontness differences result from a combination of (transla-tional) jaw fronting for /z // and rclative retraction of thc tongue body for /l/. [Supported by ESPRIT ACCOR II BRA action 3279 and NICDC Gram No. DC-00016.] 2SP14. Effects of emphatie stress and speaking modę on articulatory organization. H. Timothy Bunnell and James Polikoff (Speech Processing Lab., A. I. duPont Inst., Wilmington, DE 19899)

An adult małe taJker recorded multiple repetitions of short nonsense sentences. Each sentence was of the form “A huCl ate a C2uffle,” where Cl and C2 were from thc set {/b//d/,/g/}. Sentences containing all nine combinations of C1-C2 pairings were recorded in several speaking modes: CLEAR, CONVERSATIONAL, STRESS 1, and STRESS2.

1872 J. Acoust Soc. Am., Vol. 89, No. 4, Pt. 2, April 1991

The lattcr two conditions entailed production of the sentences in con-versational modę, but with emphatie stress on the syllable containing either the first (STRESS1) or the second (STRESS2) variable consonant. These materials were recorded at the University of Wisconsin Microbeam facility and include both acoustic data and tracings of the trajectories of flesh points on the tongue, lips, and jaw. Analyses of the acoustic and articulatory data suggest that local effects of stress on consonant articulation are simiiar to thosc of elear speech. The largest acoustic and articulatory differences were observed for consonants in syllable finał position. The various significant articulatory effects of stress and speaking modę will be compared to perceptual measures of consonant intelligibility.

2SP15. Mechanisms for tongue-tip trills. Richard S. McGowan (Haskins Labs., 270 Crown St., New Haven, CT 06511)

Possible mechanisms for the production of tongue-tip trills are discussed. It Ls neccssary that energy be transferred from the airstream to the tongue tip to enable it to vibrate. A two-mass model, simiiar to that of the vocal folds, is one possible model for tongue trills, but there are others that allow the tongue one-degree-of-freedom motion. It is possible that the boundary-layer separation point Ls affected by the acceleration of air, which is not accounted for in the quasistatic approximation. An argument will be given that the acceleration term produces a favor-able energy transfer during trills. Also, the capacitive loading of the vocal tract behind thc tongue constriction can providc necessary phase shifts between constriction volume velocity and intraoral pressurc for favorable energy transfer. A numerical simulation of the latter model is presented with parameters dcrived from measurements of oral-nasal volume ve!ocity and intraoral pressure during sustained trills. Some of the features of the trills are reproduced with this simulation.

2SP16. Formant trajectories in the speech of norraaily articulating and misarticulating children. Karen Forrcst, Ying-Chiao Tsao, Gary Wcismcr (Dept. of Commun. Disord. and Waisman Ctr., Univ. of Wisconsin-Madison, 1500 Highland Ave., Madison, WI 53705-2280), Mary Elbert, and Daniel A. Dinnsen (Indiana Univ., Bloomington, IN 47405)

In previous papers, it has been shown that analysLs of obstruent spectra can be a powerful way to quantify and classify pattems of mis-articulation in children. The focus on obstruent spectra fits in with the established idea that a problem with obstruent production is the pri-mary reason why a child may be classified as having an articulation problem, but the lesson from adult articulatory disorders [such as apraxia of speech: sce R. Kent and J. C. Roscnbek, J. Speech Hear. Res. 26, 231 (1983)) is that the prominence of obstruent articulatory prob-lems should not obscure the fact that much of the naturę of thc artic-ulatory deficit can be revealed by examination of vocalic formant trajectories. A formant trajectory analysis of five normal and five misarticulating children will bc reported, and the way in which the analysis may bear on the characterization of childhood articulatory/ phonological disorders will be described. [Work supported by NIH.) 2SP17. Production of finał consonant deletion in black dialect. Giną Michou, Sandra Hamlet (Dept. of Otolaryngol. and Dept. of Commun. Disord. and Sci., Wayne State Univ._t Detroit, MI), and Lewis Jones (Harper Hospital, Detroit, MI)

Deletion of certain finał plosives is a phonological characteristic of American black dialect. This study reports physiological observations on the production mechanisms for such consonant deletion. Yideofluo-roscopic (x-ray) speech data were obtained for monodialectical speak-ers of black dialect, who were among subjects serving as normal Controls in a study of head and neck can cer patients. Nonsense phrases of the form h i CVC were spoken when cued by a precisely articulated live

121st Meeting: Acoustical Society of America 1872