A Signal Detection Theorem Analysis of Native Japanese Production of American English Vowels
この論文をさがす
説明
The purpouse of this paper was to investigate the effects of identification training on the ability of native Japanese to pronounce the five American English (AE) vowels /æ/,/α/,/ʌ/,/ɔ/,/ɝ/. A production task, performed both before and after a six-week vowel identification training program, included an experimental group and a control group who produced words containing each of the five target vowels within a varied consonantal context. In a separate task, a group of native speakers of English performed an identification task that included a random sampling of the pre-and post-training vowel productions of the Japanese subjects from each of the two groups. A random sampling of vowel utterances as produced by the group of experimental Japanese talkers were then rated by three native English listeners in a two-interval, forced-choice identification task. A signal detection theorem-based correction for bias served to reveal a different pattern of production results and influenced the interpretation of the data, especially with regard to the /ʌ/ vowel. The results showed that the trained Japanese group's productions were more intelligible than the control group's productions for each of the five target AE vowels.
収録刊行物
-
- 青山社会情報研究
-
青山社会情報研究 2 51-62, 2010-12-25
出版社不明
- Tweet
キーワード
詳細情報 詳細情報について
-
- CRID
- 1390853649506873984
-
- NII論文ID
- 110008431827
-
- NII書誌ID
- AA12424269
-
- DOI
- 10.34321/12261
-
- NDL書誌ID
- 11240813
-
- ISSN
- 18837638
-
- 本文言語コード
- en
-
- データソース種別
-
- JaLC
- IRDB
- NDL
- CiNii Articles
-
- 抄録ライセンスフラグ
- 使用可