X-RAY MICROBEAM SPEECH PRODUCTION DATABASE

目前，在信息范围方面存在重大空白， 描述语音产生运动学，特别是当期望的数据 包括语言功能的表示。 扬声器样本和 在过去的研究中，话语库存是出了名的小， 方法和描述性惯例变化很大。 在 针对这些限制，设计了一个项目，其主要 目的是建立大样本X射线微束语音数据库， 结合了舌，唇， 下颌骨和软腭运动，与由此产生的声学 声压波，对于相对大量的扬声器（N = 50）， 丰富的话语和口头运动任务，以及长的记录间隔 （约18分钟/发言者）。 该数据库将统一各发言者， 运动任务和描述框架;足够准确和深入， 经受住内部和跨部门方差的严格统计审查 扬声器;并且，开放式可供其他演讲者无限使用 科学家 这种类型的资源将有助于开发 作为一般言语理论的基础， 生产;领导数据分析惯例的制定 也许最重要的是，这是第一次， 建立一个规范的，基于语义的语言描述 运动，反对发音行为与 沟通障碍可以比较。 收集的材料将 用于调查长期存在的五个焦点问题 （1）时间的范围和性质; 离散关节之间的协调，以及它们的运动和关节之间的协调。 产生的声波;（2）语音产生的控制目标， 从声学和运动学“目标”的变化来看， 语境和说话模式;（3） 发音机制的大小和形状，以及特征模式 口腔运动行为的性质和分布;（4）不同的口腔运动行为的性质和分布; 发音策略之间的发言者执行相同的讲话和口头 运动任务;和，（5）运动学之间的相似性，语音和 吞咽

There is currently a significant void in the range of information describing speech production kinematics, particularly when the desired data includes a representation of lingual function. Speaker samples and utterance inventories in past studies have been notoriously small, and methodology and descriptive conventions have been widely variable. In response to these limitations, a project has been designed whose primary aim is to develop a large-sample X-ray microbeam speech database, incorporating point-parameterized representations of lingual, labial, mandibular, and velar movements, in association with the resulting acoustic sound pressure wave, for a relatively large number of speakers (N = 50), rich set of utterances and oral motor tasks, and lengthy recording interval (ca. 18 min/speaker). The database will be uniform across speakers, in motor tasks and descriptive framework; sufficiently accurate and deep to withstand rigorous statistical scrutiny of variance, within and across speakers; and, openly-available for unlimited use by other speech scientists. A resource of this type will be useful for the development of informed hypotheses that must underlie a general theory of speech production; for leading the development of analysis conventions for data that are similar in kind; and, perhaps most importantly, for the first time establishing a normative, statistically-based description of speech movements, against which articulatory behaviors associated with communicative disorders can be compared. The collected materials will be used to investigate five focus issues that have been of long-standing concern to the field, including (1) the extent and nature of temporal coordination among discrete articulators, and between their motions and the resulting acoustic wave; (2) the control objectives of speech production, viewed in terms of variation in acoustic and kinematic "targets" across contexts and speaking modes; (3) the inter-relationship between measures of size and shape of the articulatory mechanism, and characteristic patterns of oral motor behavior; (4) the nature and distribution of distinct articulatory strategies among speakers performing the same speech and oral motor tasks; and, (5) the kinematic similarities between speech and swallowing.