Production and Perception of Prosodically Varying Inter-gestural Timing in American English Laterals.

Articulatory Phonetics;Laterals;Ultrasound;Linguistics;Humanities;Linguistics

Lin, Susan SychiQueen, Robin ;

University of Michigan

关键词: Articulatory Phonetics;    Laterals;    Ultrasound;    Linguistics;    Humanities;    Linguistics;   

Others  :  https://deepblue.lib.umich.edu/bitstream/handle/2027.42/86433/sslin_1.pdf?sequence=1&isAllowed=y

瑞士|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

Current theoretical approaches differ in their assessment of the influences of bio-mechanical and perceptual factors on speech production. This dissertation investigates these influences on the relative timing of the two gestures typically involved in the production of American English laterals: tongue tip raising and tongue dorsum backing. Gestural timing in onset laterals is of particular theoretical interest due to the divergent predictions of perceptually and physically based models. If speakers produce onset laterals primarily with the needs of their listeners in mind, tongue tip and tongue dorsum achievements should be roughly simultaneous, especially in higher prosodic positions where the informational load is high. In comparison, speaker-based approaches predict that onset laterals will be produced with tongue tip achievement preceding tongue dorsum achievement.These hypotheses were investigated in two experiments. A perception experiment manipulated the relative timing of tongue tip and tongue dorsum gestures in synthetic laterals. Results showed that listeners are faster and more accurate at identifying words as beginning with laterals when their onsets are created with synchronous tongue tip and tongue dorsum gestures. This finding suggests that American English laterals, and perhaps sonorants as a class, are easier to identify when their component gestures are produced synchronously.A production experiment investigated the productions of laterals by 11 American English speakers using ultrasound. Results showed that, in general, onset laterals are produced with gestural lag such that tongue tip achievement precedes tongue dorsum achievement, consistent with speaker-based approaches. The results also showed that some speakers systematically reduced gestural lag in sentence-initial relative to word-initial laterals, resulting in near gestural synchrony in sentence-initial position. The data from this speaker subset are consistent with perceptually motivated productions in which segments at the onset of large prosodic domains may be strengthened for the sake of the listener.As a whole, the findings in this dissertation are most consistent with a theory of speech production in which gestural coordination is largely determined by bio-mechanical factors which may be partially overridden, at least by some speakers, in ways that reflect speakers’ access to sophisticated phonetic knowledge about speech perception.