基于言语数据库的汉语音高下倾现象研究

提出了一种通过语句中声调音域低音点和次低音点求出语调短语低音线的方法,并以低音线为基准,考察从大规模言语语料库中随机抽取的汉语自然语句的语调特征。研究结果表明,与已有的汉语句末语调理论不同,汉语同样具有各种语言中普遍存在的音高下倾和重置的语调特征。低音线的下倾和重置可以出现在音步、语调短语和语句等不同的韵律单元中,表现最明显的是在语调短语这一级。实验结果支持汉语语调双线模型中高音线和低音线语言学功能不同的观点。     

汉语语调降阶的实验研究

采用设计特定声调组合的实验室语句的方法,研究了汉语普通话语调降阶的规律。结果发现,在普通话语调中,同样存在自动降阶和非自动降阶两种音高现象。在有低音介入的非自动降阶中,低音对后音节音域的影响表现在使音域上限降低,音域的下限保持不变;因此汉语中的降阶是语调高音特征的表现;低音特征声调会使低音后的高音发生正向降低和低音前高音的逆向提高,并且这两种作用是可以相互包容的。在自动降阶中,高音线的下降趋势是线性的,它受重音位置及重读程度的影响,在发音人中有较大差异。与已有的其他语言降阶研究的结果相比,汉语的降阶率不是固定的,且降阶的作用范围并不只局限于同低音相邻的音节。     

汉语自然对话音高研究

语音合成技术要表现会话的抑扬顿挫,首先就要对自然话语的音高做出研究.基于973电话对话语料,作者对1084个汉语语调短语的音域和音阶作了统计分析.发现语调短语可以根据其音域和音阶进行分类,语调短语的音域和音阶与它在话回中的位置有关.与朗读语言相比,自然对话语调短语的音高变化更大.     

不同语调条件下的声调音高实现

从调类个性、句中位置和重音级别3个层面的语音分析,考察普通话4个声调在不同语调条件下的音高实现。目标词被置于3种不同的焦点位置(即句重音最强的位置)和两种不同的非焦点位置(即非句重音位置)上,对目标词的调域以及目标声调的高音点和低音点进行了观察分析。实验结果表明,(1)在焦点条件以及非焦点条件下,阳平的音高位于调域的中低音区,去声低音点的理论调值尽管低于阳平低音点,但去声低音点在音高实现上往往接近阳平低音点甚至会高于阳平低音点;(2)焦点在句首位置表现为调域向上下两个方向扩展,在句末位置则表现为调域整体上抬,但不同声调的高音点并不都与调域上限同比例变化,不同声调低音点的变化也并不都与调域下限同比例变化;(3)重音后音节的音高对焦点音节的依赖关系受音步组合关系的制约,焦点和焦点后音节若在同一音步内,焦点后音节的音高与焦点音节的音高关系类似轻声音节与其前接非轻声音节的音高关系,焦点和焦点后音节之间如果存在音步边界,焦点后音节的音高表现出一定的独立性。这些结果说明了语句中声调音高实现的复杂性,一个具有较好预测性的汉语普通话语调模型的建立需要包括焦点结构、韵律结构、协同发音、调类个性等不同层面信息的诸多细节化规则。     

不同语调条件下的声调音高实现

从调类个性、句中位置和重音级别3个层面的语音分析,考察普通话4个声调在不同语调条件下的音高实现。目标词被置于3种不同的焦点位置(即句重音最强的位置)和两种不同的非焦点位置(即非句重音位置)上,对目标词的调域以及目标声调的高音点和低音点进行了观察分析。实验结果表明,(1)在焦点条件以及非焦点条件下,阳平的音高位于调域的中低音区,去声低音点的理论调值尽管低于阳平低音点,但去声低音点在音高实现上往往接近阳平低音点甚至会高于阳平低音点;(2)焦点在句首位置表现为调域向上下两个方向扩展,在句末位置则表现为调域整体上抬,但不同声调的高音点并不都与调域上限同比例变化,不同声调低音点的变化也并不都与调域下限同比例变化;(3)重音后音节的音高对焦点音节的依赖关系受音步组合关系的制约,焦点和焦点后音节若在同一音步内,焦点后音节的音高与焦点音节的音高关系类似轻声音节与其前接非轻声音节的音高关系,焦点和焦点后音节之间如果存在音步边界,焦点后音节的音高表现出一定的独立性。这些结果说明了语句中声调音高实现的复杂性,一个具有较好预测性的汉语普通话语调模型的建立需要包括焦点结构、韵律结构、协同发音、调类个性等不同层面信息的诸多细节化规则。     

汉语语句中重读音节音高变化模式研究

对汉语重读音节知觉的音高线索及句中重读音节的音高变化模式进行了研究。论文分3部分:重音知觉实验、问答匹配实验和语料库分析。重音知觉实验主要考察了重音知觉的音高线索,主要是高音点、低音点对重音知觉的贡献。重读音节音高变化模式的研究,一方面从发音人的角度,用问答匹配实验,选取/DAO4/为代表音节,设计少量实验句请多位发音人郎读,系统安排/DAO4/在句中的位置,用问句自然地引导/DAO4/重读或非重读,对这两种情况做比较;另一方面从听者的角度,用语料库分析,对一个大规模语料库通过感知实验进行重音和停顿两方面韵律标注,比较标为重和标为轻的音节的音高值。重音知觉实验结果表明,音域平移和高音点提高都是重音知觉的线索,但是高音点的提高对词重音知觉的作用更明显。重读音节音高变化模式的两项研究表明,重读音节的音高在高音线－低音线渐降汉语语调模式上变化,高音点的提高是重读音节音高变化的主要声学表现,低音点的变化更多地受到低音线渐降的限制,变化的幅度不十分明显,而且不足必须提高。高音线－低音线双线语调模型中,高音线起落的变化,前后音节高音点的对比关系表明句中音节的重读程度。     

汉语口语与朗读话语陈述句音高比较

汉语语句通常存在音高下倾现象,然而关于语句内部韵律词的具体音高表现目前的研究尚较欠缺。本研究使用的对话语料选自973电话语料库,包括69段对话,涉及79位说话人;朗读话语语料为广播电台两位主持人的新闻播音,长度为221个语句,对语句内部韵律词的高音点、低音点及音域进行了分析,结果显示对话与朗读话语多数语句的音高呈前高后低的走势,不过口语对话较长语句前半段的音高下降趋势不太明显。与朗读话语相比,口语对话韵律词的音域通常比较小。对话语句最后一个韵律词的音域相对较大,而朗读话语内部韵律词的音域大多没有差异。本研究的结果,将有助于语音合成中语句内部韵律词音阶及音域的构拟。     

汉语语句重音对音高和音长的影响

提高汉语合成语音的自然度的关键是要建立一个完善的汉语韵律模型.本文以连续的广播语言为研究对象,对汉语中语句重音对韵律特征参数的影响进行了初步探讨,分析了不同语句重音条件下音长和音高的变化及其相互关系,指出:(1)音高是语句重音的基本表达手段,随着语句重音级别的提高,音高分布曲线向高频方向推移。(2)在连续语流中词被\     

汉语和英语音高重音自动标注方法的对比与分析

虽然汉语和英语的重音自动标注被广泛的研究,但是关于汉语和英语的重音自动标注之间对比的研究还鲜有报道。基于汉语韵律标注库ASCCD和英语韵律标注库Boston University Radio News Corpus,对汉语和英语的重音自动标注的异同进行对比,考察不同的特征在不同语言的语料库上的泛化性能。通过基于集成分类回归树的重音自动标注实验、特征分析及基于互信息的重音自动标注的声学对比,得到如下结论:在相同的条件下,汉语重音自动标注的正确率比英语重音自动标注的正确率要低;在重音自动标注中,词典语法相关特征比声学相关的特征更重要;不同的声学信息源在重音自动标注中所起的作用不同,时长相关的特征对汉语和英语重音自动标注都很重要;英语中大部分特征提供的互信息要比汉语相应的特征提供的互信息要高。     

汉语普通话调制函数的实验研究

通过测量传输通路的调制转移函数而导出语言传输指数可以客观评价厅堂或扩声系统的语言清晰度。本文介绍了测得的汉语普通话的调制转移函数,比较了语言传输指数和快速语言传输指数的相关性并绘出汉语普通话音节清晰度和语言传输指数的关系曲线。     

Experimental study on declination in Chinese intonation

The aim of designing sentences with special tone combinations is to investigate the declination tendency of intonation in Chinese Putonghua.The result shows that baselines of prosodic phrases(PP),as the basic declination units,show the phenomena of declination clearly,and the declination slope is in inverse proportion to the length of PP.Different tone combinations of PP have different declination slopes.The primary conclusions are as follows: (1) If low points are at the beginning of prosodic word(PW),the absolute value of baseline slope is larger than those at the end;(2) The declination of PP is larger,if it is at the beginning of sentences than that at the end;(3) When there are several PPs in sentences,although in different syntactic relations the degree of declination is different,they have the common ground of PP declination.     

Experimental study on downstep in Chinese intonation

Downstep in pitch contour of Chinese Putonghua is examined using subtly designed sentences by controlling tone combinations. The results show both automatic and nonautomatic downstep phenomena exist in Chinese. In non-automatic downstep, low tones compress downwards the pitch range of the following syllables. and the main influence of downstep is on topline. Low tone not only lower the topline behind it, but also raise the high tones before it, the effects are compatible with each other. In automatic downstep, the topline of pitch contour in intonational phrase is presented as a linear downtrend, but it differs among speakers due to the effect of personal stress practice. In comparison with downstep phenomenon in other tone or non-tone languages, the downstep ratio in Chinese is not constant, and the domain of downstep is not limited within the adjacent tones.     

Perception of intonation in Mandarin Chinese

There is a tendency across languages to use a rising pitch contour to convey question intonation and a falling pitch contour to convey a statement. In a lexical tone language such as Mandarin Chinese, rising and falling pitch contours are also used to differentiate lexical meaning. How, then, does the multiplexing of the F(0) channel affect the perception of question and statement intonation in a lexical tone language? This study investigated the effects of lexical tones and focus on the perception of intonation in Mandarin Chinese. The results show that lexical tones and focus impact the perception of sentence intonation. Question intonation was easier for native speakers to identify on a sentence with a final falling tone and more difficult to identify on a sentence with a final rising tone, suggesting that tone identification intervenes in the mapping of F(0) contours to intonational categories and that tone and intonation interact at the phonological level. In contrast, there is no evidence that the interaction between focus and intonation goes beyond the psychoacoustic level. The results provide insights that will be useful for further research on tone and intonation interactions in both acoustic modeling studies and neurobiological studies.     

Sentence intonation in textual context--supplementary data

The experiment presented here is a sequel to a previous investigation [N. G. Thorsen, J. Acoust. Soc. Am. 77, 1205-1216 (1985)] and was designed to show whether a sequential fundamental frequency lowering of individual sentence components is present in a semantically, but not necessarily syntactically, coherent sequence (a text), when the number of sentences exceeds three. The results show that such a sequential lowering may appear, though it is not evenly distributed across the text. However, the textual intonation contour is sensitive not only to the number of sentences that make up the text, but also to the length of individual sentence components.     

Effect of intonation on cantonese lexical tones

In tonal languages, there are potential conflicts between the FO-based changes due to the coexistence of intonation and lexical tones. In the present study, the interaction of tone and intonation in Cantonese was examined using acoustic and perceptual analyses. The acoustic patterns of tones at the initial, medial, and final positions of questions and statements were measured. Results showed that intonation affects both the FO level and contour, while the duration of the six tones varied as a function of positions within intonation contexts. All six tones at the final position of questions showed rising FO contour, regardless of their canonical form. Listeners were overall more accurate in the identification of tones presented within the original carrier than of the same tones in isolation. However, a large proportion of tones 33, 21, 23, and 22 at the final position of questions were misperceived as tone 25 both within the original carrier and as isolated words. These results suggest that although the intonation context provided cues for correct tone identification, the intonation-induced changes in FO contour cannot always be perceptually compensated for, resulting in some erroneous perception of the identity of Cantonese tone.