Acoustic characteristics of American English /ɹ/ and /l/ produced by Japanese adults and children

Abstract

This study investigated the acoustic properties of American English /ɹ/ and /l/ produced by native Japanese (NJ) and native English (NE) speakers. The purpose of this study was to examine the differences in production reported in Aoyama et al. (2004) acoustically. Aoyama et al. evaluated productions of /ɹ/ and /l/ in 64 NE and NJ adults and children (16 participants each in 4 groups) using intelligibility ratings. The data were collected twice to study the acquisition of English by the NJ adults and children. In this study, four acoustic parameters (duration, F1, F2, and F3) were measured in 256 tokens each of /ɹ/ and /l/. The results showed that some aspects of acoustic parameters changed significantly over the course of one year in NJ speakers’ productions, and that the critical acoustic parameter, the F3 frequency, decreased from Time 1 to Time 2 in both NJ adults and children. However, almost all of the acoustic parameters differed significantly between NJ and NE speakers at both times of testing. The current study suggests that the perceived improvements of the NJ children’s /ɹ/ in Aoyama et al. were likely due to a combination of changes in their productions of both /l/ and /ɹ/.