Perceived foreign accentedness: Acoustic distances and lexical properties

Abstract

In this study, we examined speaker-dependent (acoustic) and speaker-independent (lexical) linguistic influences on perceived foreign accentedness. Accentedness ratings assigned to Chinese-accented English words were analyzed, taking accentedness as a continuum. The speakerdependent variables were included as acoustic distances, measured in relation to typical native-speaker values. The speakerindependent variable measures were related to the properties of individual words, not influenced by the speech signal. To the best of the authors' knowledge, this represents the first attempt to examine speaker-dependent and speakerindependent variables simultaneously. The model indicated that the perception of accentedness is affected by both acoustic goodness of fit and lexical properties. The results are discussed in terms of matching variability in the input to multidimensional representations.

Modelling the Interplay of Multiple Cues in Prosodic Focus Marking

“…In recent years, generalized additive modelling (GAM) has gained popularity and has been applied to a variety of different linguistic data such as modelling reaction times (Baayen, 2010), event-related potentials (Tremblay & Newman, 2015), accentedness ratings (Porretta et al, 2015) and dialectal variation (Wieling et al, 2014), among others. In addition to this, GAM has been extensively used in ecology (see Zuur et al, 2009).…”

The influence of gradient foreign accentedness and listener experience on word recognition

“…The results show that gradiently foreign-accented words result in differences in the strength with which they prime the intended lexical item. Previous work (Munro, 1993;Porretta et al, 2015) has shown that acoustic dimensions, taken as distances from typical native productions, influence ratings of foreign accentedness. This has been understood as a goodness-of-fit judgment between the token and the representation of an acceptable native production.…”

“…In addition to these, List (i.e., counterbalance with four levels), Trial, and Log frequency of the target word (from COCA, Davies, 2008) were included, along with random effects for Subject and Item. As stated above, mean accentedness ratings for each item were obtained from Porretta et al (2015). These ratings served as a measure of accentedness associated not only with each talker but with each experimental item allowing for a more fine-grained investigation of the effect of accentedness.…”

“…Individual sound files for each word by talker were extracted and then normalized for amplitude. In a previous study of perceived foreign accentedness (Porretta et al, 2015), ratings for foreign accentedness were collected for these tokens. In that study, 30 native English-speaking participants rated tokens from 10 talkers, which included the four talkers utilized in the present study.…”

“…Specifically, we ask whether the variability inherent to foreignaccented speech gradiently modulates activation of lexical representations, as shown through lexical priming,and gradiently affects the time-course of online spoken word recognition, as shown through visual world/word eye-tracking. Here, gradient foreign accentedness has been operationalized by means of accentedness ratings assigned previously to the auditory tokens used in the present study (see Porretta et al, 2015).…”