Control of speaking rate is achieved by switching between qualitatively distinct cognitive “gaits”: Evidence from simulation.

Abstract

That speakers can vary their speaking rate is evident, but how they accomplish this has hardly been studied. Consider this analogy: When walking, speed can be continuously increased, within limits, but to speed up further, humans must run. Are there multiple qualitatively distinct speech "gaits" that resemble walking and running? Or is control achieved by continuous modulation of a single gait? This study investigates these possibilities through simulations of a new connectionist computational model of the cognitive process of speech production, EPONA, that borrows from Dell, Burger, and Svec's (1997) model. The model has parameters that can be adjusted to fit the temporal characteristics of speech at different speaking rates. We trained the model on a corpus of disyllabic Dutch words produced at different speaking rates. During training, different clusters of parameter values (regimes) were identified for different speaking rates. In a 1-gait system, the regimes used to achieve fast and slow speech are qualitatively similar, but quantitatively different. In a multiple gait system, there is no linear relationship between the parameter settings associated with each gait, resulting in an abrupt shift in parameter values to move from speaking slowly to speaking fast. After training, the model achieved good fits in all three speaking rates. The parameter settings associated with each speaking rate were not linearly related, suggesting the presence of cognitive gaits. Thus, we provide the first computationally explicit account of the ability to modulate the speech production system to achieve different speaking styles.Speaking is a uniquely human behavior. It is by nature temporal: concepts and ideas are encoded as a stream of rapidly fluctuating sound, and the correct ordering and duration of the components is of crucial importance for intelligibility and conveying meaning. At the same time, there is great variability in the timing of speech sounds: different speakers have different habitual speech rates, and

Fluency in Dialogue: Turn‐Taking Behavior Shapes Perceived Fluency in Native and Nonnative Speech

“…This would mean that the perception of nonnative speakers' fluency is based on criteria similar to those used in fluency perception of native speakers. However, the fact that speech rate is weighed similarly in fluency evaluation tasks does not mean that the speech rate of native and nonnative speakers is processed similarly in online speech perception (e.g., Bosker & Reinisch, 2015; Kaufeld, Ravenschlag, Meyer, Martin, & Bosker, 2020; Maslowski, Meyer, & Bosker, 2019), nor that it is achieved by means of similar underlying cognitive regimes in speech production (Rodd et al., 2020).…”

A tool for efficient and accurate segmentation of speech data: announcing POnSS

“…In this article, we discuss POnSS (Pipeline for Online Speech Segmentation), a system we have created and used for segmentation work for a number of recent studies involving large-scale segmentation (Rodd et al, 2019a(Rodd et al, , 2020. With POnSS, we sought to improve the efficiency of the word segmentation task for human annotators.…”

Gait change in tongue movement

“…These differences can have profound effects on production and perception. Observations such as these have made researchers draw analogies between speech and locomotive gait: Speech simulation research 28 , corroborated by a study of reaction latency 29 , predicts that speech should be associated with differing fast and non-fast speech ‘gaits’. However, until now, such speech gaits have not been directly observed in the speech articulators.…”