母語中文者的感知訓練：英語重音的聲學線索

Perceptual Learning of English Stress by Mandarin Speakers

沈洛婷(Lo-Ting Shen)

指導教授 ： 盧郁安

國立交通大學/人文社會學院/外國語文學系外國文學與語言學研究所/碩士(2019年)

摘要

英語重音音節的聲學訊息為較高的音頻(pitch)、較強的音強(intensity)以及較長的音長(duration)。因中文為聲調語言，使用音頻高低(pitch height)及音頻輪廓(pitch contour)來分辨語意，故使中文母語使用者在判斷英語重音位置時，過度依賴音頻聲學線索(e.g., Chao 1965; Cheng 1973)，導致當重音被語調覆蓋時而誤判英語重音的位置(Ou 2010)。此論文使用感知訓練來探討中文母語使用者是否有辦法透過學習其他聲學線索，也就是音長和音強，來幫助判別重音位置。此論文包含了兩個強制性選取實驗，實驗一的訓練屏除音頻這個聲學線索，將音頻歧異化(ambiguous)，只讓中文使用者接受重音之音長及音強之訓練，而實驗二的刺激項則是各將音頻、音長或音強歧異化，使訓練的聲學訊息單一化，讓中文使用者只接受音長或音強之訓練，進而得知中文母語使用者是否有辦法透過學習其他聲學線索來幫助判別重音位置。
實驗結果比較感知訓練前測及後測重音辦別正確率發現，透過感知訓練，中文母語使用者能顯著的利用音強來增加對重音判斷的正確率，受過感知訓練之受測者比未受感知訓練之受測者皆有顯著的進步，表示利用感知訓練的方式，受試者有辦法透過這些聲學訊息幫助判別重音位置。實驗結果也呼應之前的研究發現，對於重音位置的判別，直述句的表現比疑問句好(Ou 2010)、重音位於第二音節時，相較於重音位於第一音節，受測者表現更好(Chrabaszs 2014)，且對於聲學訊息之敏銳度依序為，音頻、音強、音長(Chrabaszs 2014)。此外，本論文發現當音頻被歧異化時，中文使用者還可利用音強幫助判斷重音的位置，若用於第二語言學習上，應能透過音頻和音強來輔助中文母語使用者對於重音的學習和判斷。

關鍵字

感知訓練 ； 聲學線索 ； 英語重音 ； 非母語人士的英語重音感知

並列摘要

English stressed syllables are signified by multiple phonetic cues such as higher pitch, greater intensity and longer duration. In a lexical tone language, such as Mandarin Chinese, pitch height and pitch contour are used to distinguish different lexical meanings. It has been found that Mandarin native speakers rely mainly on pitch cues to identify English stress. (e.g., Chao 1965; Cheng 1973). As such, Mandarin native speakers have difficulty in identifying word stress when the pitch of the stressed syllables is overriden by intonation pitch contour.
This thesis investigates whether perceptual training of the other two cues, intensity and duration, may enhence the accuracy of stress identification in question intonation. If so, which cue is more efficient in the training?
One hundred Mandarin native participants were recruited. They were randomly divided into five groups. The first group received training from stimuli whose pitch cues of stress were made ambiguous. The participants were forced to turn to intensity and duration information to identify stress. The second and third groups received training from stimuli whose pitch and intensity cues, and pitch and duration cues of stress were made ambiguous, respectivly, so that the participants were forced to turn to intensity (group 2) or duration cues (group 3) to identify stress. The fourth group received training from naturally produced stimuli while the last group received no training as a control.
The results comparing the pre-test and post-test of each training condition showed that, perceptual learning can indeed assist Mandarin speakers identify English stress more accurately when the pitch information is made abscured by different intonations. The results also confirmed previous findings that, first, Mandarin speakers identify stress carried by statement intonation more successfully than that by question intonation, where the pitch information is made abscured (Ou 2010). Second, Mandarin speakers rely on pitch, and then intensity followed by duration cues in stress identification (Chrabaszs 2014). Third, Mandarin speakers identify iambic stress more successfully than trochaic one (Chrabaszs 2014). The results have implications for second language learning and pedegogy.

並列關鍵字

perceptual learning ； acoustic cues ； English stress ； non-native speakers’ perception of English stress

參考文獻

Bell‐Berti, F, Sheila Regan, and Mary Boyle. "Final Lengthening: Speaking Rate Effects." The Journal of the Acoustical Society of America 90, no. 4 (1991): 2311-11.

Boersma, Paul, and David Weenink. "Praat: Doing Phonetics by Computer (Version 5.1. 05)[Computer Program]. Retrieved May 1, 2009." 2009.

Bradlow, Ann R, David B Pisoni, Reiko Akahane-Yamada, and Yoh’ichi Tohkura. "Training Japanese Listeners to Identify English/R/and/L: Iv. Some Effects of Perceptual Learning on Speech Production." The Journal of the Acoustical Society of America 101, no. 4 (1997): 2299-310.

Chao, Yuen Ren. A Grammar of Spoken Chinese. Univ of California Press, 1965.

Cheng, Chin-Chuan. "English Stresses and Chinese Tones in Chinese Sentences." Phonetica 18, no. 2 (1968): 77-88.