Connor Mayer

The labial viseme reconsidered: Evidence from production and perception.

Previous studies have demonstrated that the labial stops /p,b,m/ may be impossible to discriminate visually, leading to their designation as a single viseme. This perceptual limitation has engendered the belief that there are no visible differences in the production of /p,b,m/, with consequences for research in machine recognition, where production differences below the level of the viseme have been largely ignored. Kinematic studies using high‐speed cine, however, have previously documented systematic differences in the production of labial consonants. This study examines the degree to which visual /p,b,m/ are discriminable in production and perception. Two experiments—one designed to measure kinematic orofacial movement using optical flow analysis and one designed to test perceiver discrimination of /p,b,m/—were used to establish the absence/presence of systematic visual differences in bilabial productions, and to replicate the previous perception findings. Results from the optical flow analysis indicat...

Talking while Chewing: Speaker Response to Natural Perturbation of Speech

This study looks at how the conflicting goals of chewing and speech production are reconciled by examining the acoustic and articulatory output of talking while chewing. We consider chewing to be a type of perturbation with regard to speech production, but with some important differences. Ultrasound and acoustic measurements were made while participants chewed gum and produced various utterances containing the sounds /s/, /ʃ/, and /r/. Results show a great deal of individual variation in articulation and acoustics between speakers, but consistent productions and maintenance of relative acoustic distances within speakers. Although chewing interfered with speech production, and this interference manifested itself in a variety of ways across speakers, the objectives of speech production were indirectly achieved within the constraints and variability introduced by individual chewing strategies.

A Challenge for Tier-Based Strict Locality from Uyghur Backness Harmony

In this paper we describe the process of backness harmony in Uyghur, where suffix forms are determined first from the backness of certain vowels in the stem, or, if no such vowels are present, from the backness of dorsals in the stem. We show that this pattern cannot be captured by a tier-based strictly local (TSL) language. This is problematic for the weak subregular hypothesis, which claims that all segmental phonological stringsets are TSL languages. Next, we consider an alternative phonological analysis that is compatible with a TSL representation, but empirically unsupported. Finally, we consider the possibility that Uyghur backness harmony might be a lexicalized pattern, and find some suggestive evidence in support of this. This alternative appears to be the most likely way in which Uyghur backness harmony might, in principle, turn out to be compatible with the hypothesis that TSL languages provide an upper bound on phonological learnability.