句法歧义的韵律声学表现初探

发布时间：2018-05-04 08:05

本文选题：韵律特征 + 句法歧义　；参考：《上海外国语大学》2013年硕士论文

【摘要】：本文运用实验语音学的方法，从讲话者角度研究了四类句法歧义消歧时的韵律表现。研究的对象包括动宾/偏正类结构歧义、并列/偏正类结构歧义和偏正、主谓类层次切分歧义。笔者从现有文献中精心挑选各类型歧义短语若干个，将它们放置在不同的上下文中，请四位长期使用普通话的发音人在自然状态下进行朗读。然后逐个分析歧义短语在不同语境下的韵律特征，并运用语音学相关理论进行分析解释。研究发现：1.影响歧义短语内部各音节相对时长的因素有很多种，包括歧义短语本身包含的音节数，歧义短语在句中的位置，短语中各音节的声调以及各音节包含的音素数等。这些因素决定了特定歧义短语内部音节之间的相对时长关系，且这种时长关系不会因为发音人采用时长策略消歧而轻易改变。2.动宾/偏正类结构歧义在表达不同语义时表现出不同的时长特征，但这种时长表现还不具有显著规律，只能说是具有一种倾向性。音高特征对这类歧义的消歧作用更有效一些，主要手段是在语法重音或语义焦点上扩大调域和抬高基频。相比之下，时长和音高特征对并列/偏正类结构歧义的消歧作用比较明显，但不同发音人之间并没有表现出非常一致的目标点。3.对由不同层次切分可能造成的歧义，发音人均能使用时长和音高特征进行有效消歧。消歧策略主要有三种：一种是延长语法层次边界前音节或前两音节的时长并在边界处设置音高重置以造成停顿感知，，从而达到在连续语流中形成层次切分的目的。另一种是通过提示语法重音来达到在语流中形成层次切分的目的，主要手段是延长某些语法重音的时长或抬高基频。第三种策略是发音人根据自己的理解强调语义焦点，从而使歧义句的不同语义选择得以区分。
[Abstract]:This paper studies the prosody performance of four kinds of syntactic ambiguity disambiguation from the perspective of the speaker by means of experimental phonetics. The research objects include verb-object / positive-class structure ambiguity, paratactic / positive-class structure ambiguity and bias, subject-predicate class hierarchy segmentation ambiguity. The author carefully selects a number of ambiguous phrases from the existing literature and places them in different contexts. Four speakers who have been using Putonghua for a long time are asked to read aloud in a natural state. Then, the prosodic characteristics of ambiguous phrases in different contexts are analyzed one by one, and the relevant theories of phonetics are used to explain them. The study found that 1: 1. There are many factors affecting the relative duration of each syllable in the ambiguous phrase, including the number of syllables contained in the ambiguous phrase, the position of the ambiguous phrase in the sentence, the tone of each syllable in the phrase and the phonological prime number contained in each syllable. These factors determine the relative length of time between syllables in a given ambiguous phrase, and this relationship will not change easily because the speaker adopts a time-length strategy to disambiguate. Verb-object / positive-class ambiguity shows different characteristics of time length when expressing different semantics, but this kind of long expression has not a significant rule, but it can only be said to have a tendency. Pitch features play a more effective role in disambiguation of this kind of ambiguity. The main means are to expand the tone domain and raise the fundamental frequency in the grammatical stress or semantic focus. In contrast, the disambiguation effect of time length and pitch on paratactic / orthotropic structure ambiguity is obvious, but there is no very consistent target point. 3. For possible ambiguity caused by different levels of segmentation, the speaker can effectively disambiguate by using duration and pitch features. There are three main disambiguation strategies: one is to extend the duration of the preceding syllable or the first two syllables at the grammatical level boundary and to set the pitch reset at the boundary to cause pause perception so as to achieve the purpose of hierarchical segmentation in the continuous speech flow. The other is to achieve the purpose of hierarchical segmentation in the flow by prompting the grammatical stress. The main means is to extend the duration of some grammatical stresses or to raise the fundamental frequency. The third strategy is to emphasize the semantic focus according to one's own understanding, so that the different semantic choices of ambiguous sentences can be distinguished.
【学位授予单位】：上海外国语大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：H116;H146.3

【参考文献】