Gestural and prosodic influences on the articulation of consonant sequences
Dani Byrd
Abstract
Dani Byrd
Abstract
This study investigates several factors affecting articulatory timing and reduction in English consonant sequences. Electropalatography was used to record five speakers saying sentences containing a consonant cluster spanning a word boundary. Based on the EPG contact profiles, metrics were calculated indicating the overlap of the consonants and their spatial displacement, i.e., magnitude. The following results were obtained. (1) Coronal consonants are more overlapped by a following velar stop than a velar stop is by a following coronal. (2) A stop is more overlapped by a following consonant than is a fricative. (3) Movements in coda position are smaller than movements in onset position for stops but not for fricatives. (4) Movements in coda position are more variable than movements in onset position for stops but not for fricatives. Findings (3) and (4) suggest that variability may play an important role in reduction processes. In summary, these findings support the claim that speech coordination is affected by linguistic factors. These factors include both gestural influences—such as place and manner of articulation—and prosodic influences—such as syllablic position. This experiment provides significant data with which models of speech timing can be further explicated to more accurately reflect articulatory coproduction. [Work supported by NSF.]
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
This study investigates several factors affecting articulatory timing and reduction in English consonant sequences. Electropalatography was used to record five speakers saying sentences containing a consonant cluster spanning a word boundary. Based on the EPG contact profiles, metrics were calculated indicating the overlap of the consonants and their spatial displacement, i.e., magnitude. The following results were obtained. (1) Coronal consonants are more overlapped by a following velar stop than a velar stop is by a following coronal. (2) A stop is more overlapped by a following consonant than is a fricative. (3) Movements in coda position are smaller than movements in onset position for stops but not for fricatives. (4) Movements in coda position are more variable than movements in onset position for stops but not for fricatives. Findings (3) and (4) suggest that variability may play an important role in reduction processes. In summary, these findings support the claim that speech coordination is affected by linguistic factors. These factors include both gestural influences—such as place and manner of articulation—and prosodic influences—such as syllablic position. This experiment provides significant data with which models of speech timing can be further explicated to more accurately reflect articulatory coproduction. [Work supported by NSF.]
Key concepts: Coda, Consonant, Place of articulation, Articulation (sociology), Consonant cluster, Stop consonant, Speech recognition, Position (finance)