The role of production processes in determining word duration

我们说话的节奏和时机在交流中起着重要的作用。例如，说话者往往在句子中更快地说出9这个词。在时间上缝一针比在句子中说9更快，我想到的数字是9。时间上的一些差异是由我们希望帮助听者理解我们的愿望造成的：与第一句相比，第二句中的单词9更难预测，所以我们延长了它，让我们的听者有额外的时间来理解它。或者，说话者可以在第一句话中缩短九个，因为这样说起来更容易。在说话过程中减速和提速可能会给生产系统提供所需的时间来计划和产生准确、连续的语音。这个项目考察了在演讲过程中发生的计划的类型，以及它们可能影响我们的语速的不同方式。这个项目的结果可能最终会导致对言语障碍患者进行更好的干预，并在设计听起来更自然的人工语音系统方面取得进展。语言研究人员面临的一个难题是理解影响说话者如何实现特定单词的因素。例如，给定词的长度至少部分取决于该词在话语中的信息量(例如，它有多可预测，它提供了多少新信息，等等)。人们不太了解的是，每个单词的时间和持续时间背后的具体因素和机制。这个项目测试了一种假设，即不同水平的语言规划对发音时间和单词时长有不同的影响，研究人员将这一理论称为双域假说。在这一假设下，词汇和概念规划影响开始发音的时间，而与语音顺序相关的规划过程影响产生单词本身的时间。初步研究表明，当一个单词的发音顺序编码过程很复杂时，这个单词会被加长。另一方面，当这种语音编码顺利进行时，单词就会变短。初步实验还表明，词汇启动对S发音的开始时间有影响，但对S的发音时长没有影响。该实验使用事件描述范式来检验双域假设的预测，并将导致更好地理解计划的哪些方面影响单词持续时间。

The rhythm and timing of our speech plays an important role in communication. For example, speakers tend to say the word nine more quickly in the sentence A stitch in time saves nine than in the sentence The number I am thinking of is nine. Some of these differences in timing are driven by our desire to help the listener understand us: the word nine is less predictable in the second sentence than it is in the first, so we lengthen it to give our listeners extra time to understand it. Alternatively, speakers may shorten nine in the first sentence because it is easier to say. Slowing down and speeding up during speaking may give the production system the time it needs to plan and produce accurate, continuous speech. This project examines the types of planning that occur during speech and the different ways they may affect our speaking rate. The results of this project may eventually lead to better interventions for individuals with speech disorders and to advances in designing artificial speech systems that sound more natural.A puzzle for language researchers is understanding the factors that affect how speakers actualize specific words. For example, the length of a given word is driven, at least in part, by how informative the word is in the discourse (e.g., how predictable it is, how much new information it provides, etc?). What is less well understood are the specific factors and mechanisms that underlie the timing and duration of each word. This project tests the hypothesis that different levels of linguistic planning have differential effects on time to articulation and word duration, a theory the investigator calls the ?two-domain hypothesis.? Under this hypothesis, lexical and conceptual planning influence the time to begin articulating a word, while planning processes related to phonological ordering affect the timing of producing the word itself. Preliminary work suggests that when the process of encoding the order of a word?s sounds is complex, the word is lengthened. On the other hand, when this phonological encoding proceeds smoothly, the word is shorter. Preliminary experiments also suggest that lexical priming affects the time to begin a word?s articulation, but not the word?s duration. The proposed experiments use an event-description paradigm to test the predictions of the two-domain hypothesis and will lead to a better understanding of which aspects of planning influence word duration.