新竹地區台灣閩南語海口腔聲調的聲學研究

An acoustic study of tones in Hsinchu Coastal Taiwanese Southern Min

彭尹鴻(Yin-hung Peng)

指導教授 ： 張月琴 ； 謝豐帆

國立清華大學/人文社會學院/語言學研究所/碩士(2011年)

https://doi.org/10.6843/NTHU.2011.00600

摘要

This thesis provides detailed acoustic descriptions of tones of the Coastal Taiwanese Southern Min (CTSM) spoken in Hsinchu, Taiwan in terms of the acoustic properties of fundamental frequency (F0), intensity and duration. In previous studies, tones of CTSM sub-dialects have been transcribed based merely on auditory impression. This study investigates the tones of Hsinchu CTSM in an instrumental way by examining the acoustic properties of tones produced in three contexts: tones in isolation (monosyllabic tones), tones on the final syllables of disyllabic tonal sequences (final tones) and tones on the non-final syllables of disyllabic tonal sequences (sandhi tones). In this study, the tones in these three contexts are accurately described and their tone values are given. Furthermore, a comparison is made between the tone values of Hsinchu CTSM tones as proposed in Ang (1997a) and those given in this study.
The main findings are as follows: First, the tone values of final tones and monosyllabic tones proposed in this study are generally similar to those of the citation tones proposed in Ang (1997a), but some important discrepancies between the findings of these two studies can be found in terms of the tone values of tones belonging to the categories of yangping and yangru; as well as in terms of the tone values of sandhi tones belonging to the categories of yangping, yingqu, yangqu, yingru and yangru. In addition, in this study, neutralization in tone sandhi is observed among/between different sandhi tones and also between sandhi tones and monosyllabic tone. Next, as for tones in isolation and on final syllables, unchecked falling tones tend to be shorter than the other unchecked tones. Finally, by treating the pitch contours of tones in isolation as canonical forms, the methods of curve-fitting and discriminant analysis (Andruski and Costello 2004) used in this study can not only measure the degree of similarity between the pitch contours of both non-canonical forms (final/sandhi tones) and canonical forms (monosyllabic tones), but also indirectly measure the degree of similarity between the pitch contours of the non-canonical forms produced in different contexts (final tones and sandhi tones).

關鍵字

聲調 ； 變調 ； 混同 ； 曲線擬合 ； 區別分析

並列摘要

本論文試圖從基頻(fundamental frequency)、振幅(intensity) 和時長 (duration)三方面的聲學特性對新竹地區台灣閩南語海口腔方言的聲調做一個詳盡的聲學描述。前人研究對於台灣閩南語海口腔方言的聲調的描述僅憑藉於聽覺，而此研究則以儀器測量之方式來檢驗新竹地區台灣閩南語海口腔方言聲調在孤立、雙字詞後字和雙字詞前字三種不同的環境下之基頻、振幅和時長等三個聲學特性，在此三個環境下所發出的聲調在本文分別被稱為單字調、後字調和變調。對於此三種聲調，本文將會做精確的聲學描述並提供其調值。此外，我們會將洪惟仁(1997a)所提出之新竹地區閩南語海口腔聲調之調值與本研究所提出的調值做比較。
本文有幾個主要發現：第一，此研究所提出的單字調和後字調之調值大致上和洪惟仁(1997a)所提出的本調調值類似，但對於陽平本調、陽入本調、陽平變調、陰去變調、陽去變調、陰入變調和陽入變調之調值而言，兩研究之間也是有重要差別的。第二，變調現象(tone sandhi)中的混同(neutralization)情形會出現在不同變調之間、後字調與單字調之間或者變調與單字調之間。此外，我們發現在孤立和雙字詞後字兩個環境下，舒聲降調的時長會傾向比其他舒聲調的時長短。最後，如將孤立調之聲調曲線假設為該語言聲調的正規形式(canonical forms)，利用Andruski and Costello (2004) 所提出的曲線擬合法(curve-fitting)和區別分析法(discriminant analysis)，我們不只能夠測量非正規形式(後字調或變調之聲調曲線)與正規形式(單字調之聲調曲線)之間的相似度，還能夠間接測量不同環境下之非正規形式(後字調和變調之聲調曲線)之間的相似度。

並列關鍵字

tone ； tone sandhi ； neutralization ； curve fitting ； discriminant analysis

參考文獻

Hsieh, Peggy Pei-Yu. 2004. Variation in Taiwanese Tones. National Chung Cheng University dissertation

Ang, Ui-Jin. 2007. Taiwan diqu yuyan fangyan fenqu yu fenbu diao cha [A survey on the dialect distribution in Taiwan]. Taipei: National Science Council (NSC 95-2411-H-142-001).

Ang, Ui-Jin. 2008. Taiwan diqu yuyan fangyan fenqu yu fenbu diao cha [A survey on the dialect distribution in Taiwan]. Taipei: National Science Council. NSC 96-2411-H-142-005.

Ang, Ui-Jin. 1985. Taiwan Helao yu shengdiao yanjiu [Studies on the tones of Taiwanese Southern Min]. Taipei: The Zhili Press.

Ang, Ui-Jin. 1992. Taiwan Fangyan zhi Lu [A dialect trip in Taiwan]. Taipei: Vanguard Publishing House.