Comparisons of Durations of Stops with Voicing Differences Across Word Boundaries

Abstract

This study is aimed to investigate whether there are differences of durations of homorganic stops with voicing differences across word boundaries between native speakers of English and Korean. For data collection, eleven American English and twenty one Korean speakers participated in a production experiment. The participants’ sounds were recorded and analyzed with Praat, and statistics were done with SPSS. The durations of the stops produced by English speakers are in the order of /voiceless#voiceless/, /voiced#voiceless/, /voiceless#voiced/, and /voiced#voiced/. This is because durations of word-final stops and VOTs of word-initial stops are different according to their voicing features. In the Koreans’ productions, durations of /voiceless#voiceless/ and /voiced#voiceless/ are similar to each other, and /voiceless#voiced/ and /voiced#voiced/ are also similar to each other. This is because durations of word-final stops are similar to each other due to the so-called coda neutralization in Korean. Thus, the durations of word boundary stops in the Koreans’ productions are distinguished based only by the VOTs of word-initial stops. This study also illustrates some other interesting findings in detail.