885095957

ration results in a comprehensive remapping between closurc duration and phonctic calegory, or whcther trading effects are confined to the category boundary. Two serics of disyllables were created ranging from /aba/ through /apa/ to */apa/ (an cxaggcratcd /p/) having initial vowel durations of 153 and 250 ms, rcspcctively. Closurc duration in each scrics varicd from 20 to 400 ms. A preliminary experiment rc-vealcd a standard trading relation, in that the /b/-/p/ category boundary was located at a longer closure duration for the stimuli with a long. compared to a short, prcceding vowel. In the main experiment, listeners were asked to judge each disyllable in each senes for the goodness of its consonant as a membcr of the /p/ category. For both senes the /p/ category was perceived as having internal structure, with a limited rangę of stimuli bcing judged as the best exemplars. Furthermore, the rangę of best exemplars for the long vowel series was displaccd relative to that for the short vowel series toward longer values of closure duration. These findings indicate that the acoustic properties in question trade against each other not only at the phonetic category boundary but a ko within the category. This results in a comprehensive remapping of phonetic category structure similar to that observed in earlier rescarch for changes in speaking ratę. [Work supported by NIH.]

10:30

8SP10. Influence of a syllable*s form on the perceired internal structure of roking categories. Lydia E. Volaitis and Joannę L. Miller (Dept. of Psychol., Northeastem Univ., Boston, MA 02115)

The role of syllable structure on voice-onset time (VOT) was ex-amined by comparing VOT values in consonant-vowel (CV) and consonant-vowel-consonant (CVC) syllabłes, across a rangę of speaking rates. In a production study, whcn CV and CVC syllablcs were equatcd for overall duration, VOT values were found to be consistently shorter for the CYC than the CV syllabłes. Furthermore, when syllabłes were equated for CV duration, YOT values for CV and CVC syllabłes tended to be equal, suggesting that speakers were producing YOT values with regard to the syllable’s CV duration. In a subsequent perception study. listeners adjusted for these changes in VOT by altering three aspects of category structure in relation to the syllabłe’s CV duration, and not to its ovcrall duration—the location of the voiced—voicelcss category boundary, the upper limit of the voice!ess category, and the rangę of "good” exemplars that lies between those two boundaries. These findings support the notion that listeners perceptually restructure their phonetic categories so as to accommodatc changes in VOT that occur in production as a result of the syllable’s phonological context, as well as its speaking ratę. [Work supported by NIH.]

10:45

8SP11. Independence of scene analysis and the speech module. D. H.

Whalen and Alvin M. Liberman (Haskins Labs., 270 Crown St., New Haven. CT 06511)

Even the first example of the duplex effect [Rand, J. Acoust. Soc. Am. 55, 678-680 (1974)] gives evidence that speech perception can bring together portions of the speech signal that scene analysis says are separatc: Although a formant transition on one ear sounds like a non-speech “chirp," ils speedi Information is used by the syllable on the other ear. The present work explores the competition between these two organizations of the signal by extending previous work [Whalen and Liberman, Science 237, 169-171 (1987)] in which T3 formant sinusoids providcd speech information both below and above the intensity level at which the nonspeech aspect of the tonę was just perceptible. As an extension, altemative “scenes" were madę plausible for the transitional tones by (1) extending them 50 ms before the syllable; (2) extending them 50 ms after the syllable; (3) ercating two harmonics hased on the tonę; and (4) creating four harmonics based on half the frequency of the tonę. Identification of the speech sounds was always above chance, in-dicating the success of the speech module. However, the łiarmonic series (below the detection level) and the precursor tonę (below and above the detection level) reduced the accuracy of the speech judgments, suggesting competition with scene analysis. Scene analysis and speech perception, though independent, are often in competition for the same signal, and speech takes precedence in a wide array of circumstances. [Work supported by NIH Grant HD-01994.]

11:00

8SP12. Effects of lexical status on perceptual organization In duplex perception. Lynne C. Nygaard (Speech Res. Lab., Dept. of Psychol. Indiana Univ., Bloomington, IN 47405)

Duplex perception occurs when a synthetic syllable is split so the third-formant transition is presented to one ear and the rest of the syllable (the base) is presented to the other ear. Listeners report hearing two distinct percepts—a complete syllable in the ear with the base and a nonspeech chirp in the ear with the transition. A modification of this duplex phenomenon can be created by presenting a third-formant transition in isolation to one ear and the same transition electronically mixed with the base to the other ear. In this case, the transition information fuses to form a chirp percept in the center of the head and the syllable in the other ear hecom es clearer than the one produced with the standard dupiex procedurę. Nygaard [J. Acoust Soc. Am. Suppl. I 87, S71 (1990)] found that when the spectral composition or onset fre-quency of the isolated transition was varied relative to the complete syllable base, both phonetic integration and nonphonetic fusion re-mained remarkably Lntact even with large diflferences in spectral com-position between components. In a series of experiments, the lexical status of the syllable base was varied to determine the effcct of lexical information on perceptual organization of acoustic components that difter in spectral composition. It was found that the lexical status of the eventual phonetic percept infiuenced the phonetic integration of acoustic components into syllable percepts, but had no effect on fusion of the third-formant transitions to creatc a centered chirp percept. These results suggest that lexical information contributes to the perceptual grouping of acoustic components into phonetic percepts.

11:15

8SP13. An investigation of loeus equations as a source of relational invariance for stop place categorization. Harvey M. Sussman (Dept. of Linguistics and Speech Commun., Univ. of Texas, Austin, TX 78712)

Loeus equations were investigated as a potential higher-order metric capable of illustrating relational invariance for place of articulation in voiccd initial stop consonants indepcndently of vowel context. Loeus equations are straight-line regression fits to data points formed by plot-ting onsets of FI transitions along the y axis and their corrcsponding midvowel nuclei along the x axis. Twenty subjects, 10 małe and 10 female, produced /b/v/t/, /d/v/t/, and /g/v/t/ tokens for 10 vowel contexts. Each CVC token was repeated in a carrier phrase five times yielding 150 tokens per subject. Formant measures were obtained using the MacSpeech Lab II speech analysis system. Loeus equation scatter-plots revealcd extremely tight clustering of points around the regression linę that were consistent across speakers and gender. Dcrived slopc and y-intercept parameters were significantly different across stop place categories. The relative value of FZo^ as it linearly changes in relation to the coarticulatorily produced vowel reflects an acoustic correlate of relational invariance for stop place. A discriminant analysis using Flonyti a a* prcdictors showed 82%, 78%, and 67% classification rates for labial, alveolar, and velar place. Using derived slope and y-intercept values as predictors led to 100% classification into stop place categories. A neurobiologically oriented perspective on the invariance issue is explored and a brain-based recognition algorithm for stop place integrating burst and FI cues is offered. [Work supported by NSF.]

1998 J. Acoust. Soc. Am., Vol. 89, No. 4, Pt. 2, April 1991 121st Meeting: Acoustical Society of America 1998