US-Sweden Cooperative Science: Intonation and Voice Source Characteristics

美国-瑞典合作科学：语调和声源特征

• 批准号: 8712375
• 负责人: Janet Pierrehumbert
• 金额: $ 0.93万
• 依托单位: Lucent Technologies, Inc
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 1988
• 资助国家: 美国
• 起止时间: 1988-01-01 至 1989-06-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=8712375&HistoricalAwards=false
• 关键词: US; Sweden; Cooperative; Science; Intonation

This award will support collaborative research between Dr. Janet Pierrehumbert of AT&T Bell Laboratories and Professor Gunnar Fant, Royal Institute of Technology, Stockholm, in the area of linguistics. The aim of the project is to establish rules for the influence of intonation patterns in speech on the perceived characteristics of the voice source. As a first step, intonation patterns will be related to detailed characteristics of the air flow in the throat. Experiments will be carried out in which the important factors affecting intonation, such as stress pattern, choice of melody, and pitch range, are varied independently. The air flow will be analyzed non-invasively by using already-known functions to transform between the sound actually produced and air flow details. Studies will be made using both English and Swedish speakers, taking advantage of the very different intonation characteristics of these languages. The Swedish collaborator has pursued studies on the effects of speech patterns on air flow and therefore has direct experience with the measurement and use of these flows for speech characterization. The description of air flows developed by the Swedish investigator will be combined with the linguistic model of intonation developed by the U.S. investigator. This research may lead to important applications in the fields of linguistics and speech synthesis.

该奖项将支持AT&Amt Bell实验室的Janet Pierrehumbert博士和斯德哥尔摩皇家理工学院的Gunnar Fant教授在语言学领域的合作研究。该项目的目的是建立语音中语调模式对声源感知特征的影响的规则。作为第一步，语调模式将与喉咙中气流的详细特征相关。实验中，影响语调的重要因素，如重音模式、旋律的选择和音域，将独立变化。通过使用已知的函数在实际产生的声音和气流细节之间进行转换，将对气流进行非侵入性分析。将使用英语和瑞典语进行研究，利用这两种语言非常不同的语调特点。这位瑞典合作者一直在研究语音模式对气流的影响，因此在测量和使用这些气流进行语音表征方面有直接经验。瑞典调查员对气流的描述将与美国调查员开发的语调语言模型相结合。这项研究可能会在语言学和语音合成领域产生重要的应用。