|کد مقاله||کد نشریه||سال انتشار||مقاله انگلیسی||ترجمه فارسی||نسخه تمام متن|
|1100658||953472||2016||19 صفحه PDF||سفارش دهید||دانلود کنید|
• American English-speaking listeners grouped talkers by perceived L1 background.
• L1 American English talkers formed a clear perceptual category.
• L1 Hindi talkers formed a consistent subcategory of non-native talkers.
• L1 Korean and L1 Mandarin talkers were perceptually similar.
• Deviation from native phonetic norms predicted many classification responses.
Listeners׳ ability to distinguish native from non-native speech has been robustly attested, but less is known about the nature of perceptual subcategories within non-native speech. This study examined American English-speaking listeners׳ abilities to perform auditory free classification by talker native language (L1) background on the basis of CV- and word-length excerpts of English productions by L1 American English, L1 Hindi, L1 Korean, L1 Mandarin, and L1 Spanish talkers. Results are examined in terms of classification accuracy and perceptual similarity, and the phonetic properties predicting listeners׳ classification responses are explored. Overall, L1 American English talkers were grouped together, as were L1 Hindi talkers. Talkers from L1 Korean and L1 Mandarin were perceptually similar to one another, although they did not comprise a single shared group. Deviation from native norms in VOT and in spectral and temporal properties of vowels predicted listeners׳ classification responses. These results suggest that listeners can use specific phonetic properties of the speech signal to group talkers on the basis of very little input, and that L1 Hindi talkers form a clear and consistent subcategory of non-native talkers for American English-speaking listeners.
Journal: Journal of Phonetics - Volume 55, March 2016, Pages 19–37