Elsevier

Journal of Phonetics

Volume 25, Issue 2, April 1997, Pages 187-205
Journal of Phonetics

Regular Article
The extent of vowel-to-vowel coarticulation in English

https://doi.org/10.1006/jpho.1996.0041Get rights and content

Abstract

This acoustic study investigated the extent and nature of vowel-to-vowel coarticulation in English trisyllabic utterances (/bV1bəbV3b /) where both V1 and V3 were either /a/ or /i/. Primary stress was assigned to either the first or third syllable, and secondary stress was assigned to the other syllable. The study revealed considerable variability among the four speakers in the systematic effects of non-adjacent vowels on one another. The most important result is that coarticulatory effects can, in some instances, extend beyond the bounds that previous research had assumed; both carry-over and anticipatory effects extend from one full vowel, through the medial schwa, and into the midpoint of the other full vowel. In fact, for one speaker, effects of V3 can be seen as early as V1 onset of /a/. This result indicates that the foot does not define the domain over which coarticulatory effects operate. Finally, this study found no evidence of greater carry-over than anticipatory effects.

References (0)

Cited by (73)

  • Coarticulation as synchronised CV co-onset – Parallel evidence from articulation and acoustics

    2022, Journal of Phonetics
    Citation Excerpt :

    Because of this difference, the presence of acoustic or articulatory movement toward the following vowel prior to the acoustic consonantal landmark does not necessarily contradict the articulatory syllable hypothesis. In fact, based on the articulatory definition of segment onset, the finding of Öhman (1966) as well as the subsequent reports of anticipatory coarticulation (Magen, 1997; Mok, 2012; Recasens & Pallarès, 2001; Rubertus & Noiray, 2018) could be interpreted as evidence for the articulatory syllable. That is, the beginning of the formant movement toward the next vowel well before the consonant closure, as seen in Fig. 1b, would suggest that the articulation of that vowel has started at that moment.

  • Detecting anticipatory information in speech with signal chopping

    2020, Journal of Phonetics
    Citation Excerpt :

    Coarticulatory effects between overlapping gestures can be “planned”, i.e. are partly under speaker control (Whalen, 1990), and so language- and speaker-specificity of such effects does not necessarily indicate that they arise from non-local interactions. More compelling examples of non-local interactions between vowels have been observed in effects on vowels from non-adjacent syllables (Grosvald, 2009; Magen, 1997), and in paradigms where a planned but non-executed vowel has an influence on a produced vowel (Tilsen, 2007, 2009). Moreover, observations of non-local interaction are not confined to vowels.

  • Functional brain outcomes of L2 speech learning emerge during sensorimotor transformation

    2017, NeuroImage
    Citation Excerpt :

    In considering why simple contexts led to more acoustically accurate imitation of non-native items, it is important to note that both of the non-word contexts (mono and trisyllabic) afforded similarly accurate imitations of native vowels; however, only the simpler monosyllabic context facilitated improved imitations of the non-native vowel (relative to imitation of that vowel in isolation or in trisyllables). Based on our predictions, this appears to suggest that the increased articulatory complexity of the trisyllabic context, when combined with the non-native vowels, impacted participants’ ability to accurately imitate the mid-stressed /y/ vowel (further to Magen, 1997; see Kühnert and Nolan, 1999). We can then ask why isolated vowels were associated with less accurate imitations compared to in a simple context, particularly in the case of non-native /y/ (cf. context effects in perceiving French vowels; Gottfried, 1984).

View all citing articles on Scopus
View full text