어말자음의 차이를 반영한 선행모음의 길이 교정효과에 대한 연구

Abstract

It is well known that voicing features of English word-final consonants affect duration of the preceding vowels. Native speakers of English tend to produce a short vowel when it precedes a voiceless consonant, and a long vowel when it precedes a voiced consonant. Two groups of Korean students learned how to pronounce these types of words such as ‘cap’ and ‘cab’. One of the groups was trained by a native speaker of American English and the other by a Korean phonetics professor. The former repeated after the English speaker to correct their pronunciations, while the latter learned phonetic explanations about the difference of the vowel durations of these words. Then both groups practiced the words used in the experiment. After the training sessions, they recorded these words. We measured the durations of their vowels produced both before and after the training sessions. According to the acoustic analyses and statistical results, the latter teaching method was more effective than the former. Most of the words produced by the latter group had short vowels before voiceless sounds and long vowels before voiced sounds. The former group’s performance was less than that of the latter group. The acoustic results also showed that the Korean language’s phonological contrast affected their pronunciations. Especially the coda neutralization had an effect on their pronunciation of word-final stops. That is, they tend to produce unexploded stops for the word-final stops of the words used in the experiment.