Engineering Issues in Understanding Human Speech

理解人类语音的工程问题

• 批准号: 0828903
• 负责人: Michael Plesniak
• 金额: $ 22.38万
• 依托单位: Purdue University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2008
• 资助国家: 美国
• 起止时间: 2008-08-01 至 2010-05-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0828903&HistoricalAwards=false
• 关键词: Engineering; Issues; Understanding; Human; Speech

CBET-0828903M. Plesniak, Purdue UniversityUnder this grant, modeling assumptions that are commonly applied in models of human phonation (fluid-induced vibrations of the vocal folds) will be systematically evaluated. Because of the inaccessibility of the larynx, modeling has long been a primary component of voice research. Speech models seek to simplify the complicated fluid-structure interactions of human voice production while still capturing the relevant physics. However, untested modeling assumptions limit the accuracy of such models. Ultimately, highly accurate models will be transformative to clinical diagnosis, treatment, and pre-surgical planning of voice disorders. The prevalence of voice disorders in the general population is estimated to be as high as 30%, and is twice as high among professionals who lecture regularly, such as educators. Each year, billions of dollars in lost productivity costs is due to vocal ailments. The goal of this project is to assess the validity of untested modeling assumptions that are routinely employed. Fluid flow investigations will be performed in a 7.5 times life-size controllable dynamic model of the vocal folds. Viscous flow phenomena within the glottis will be studied via spatial and temporal flow field data using Particle Image Velocimetry (PIV) and Laser Doppler Velocimetry (LDV). Parametric structural analysis will be performed to determine the sensitivity of the vocal fold structure to eight geometric and six tissue parameters of several have never been measured. The knowledge obtained from this study will increase the accuracy and validity of future models, eventually leading to the creation of clinically relevant models of human speech. In addition to speech production, the results will be relevant to the more general fields of fluid-structure interactions, and biomedical engineering.The existing outreach programs at Purdue University will ensure that in addition to engineering and societal impacts, the research will have a broader impact on participation of underrepresented groups, undergraduate research participation, and K-12 outreach programs. The Summer Undergraduate Research Fellowship (SURF) program at Purdue University facilitates meaningful guided research for undergraduate students, in which female and minority participation is especially encouraged by Purdue's Minority Engineering organization. Participation in engineering outreach programs such as "Imagination Station", "Middle School Minds", and EPICS at Purdue University will facilitate the exposure of K-12 students to the excitement of engineering at a young age. A functional model of the human speech process created in conjunction with EPICS will be used for instruction in undergraduate Speech Language and Hearing Research (SLHS) classes.

CBET-0828903M。Plesniak，普渡大学在这项资助下，将系统地评估通常应用于人类发声模型（流体引起的声带振动）的建模假设。 由于喉部的不可接近性，建模长期以来一直是语音研究的主要组成部分。 语音模型试图简化人类语音产生的复杂流体-结构交互，同时仍然捕获相关的物理学。 然而，未经测试的建模假设限制了这些模型的准确性。 最终，高度准确的模型将对语音障碍的临床诊断、治疗和术前规划产生变革性影响。 据估计，普通人群中嗓音障碍的患病率高达30%，而经常讲课的专业人员（如教育工作者）的患病率是普通人群的两倍。 每年，数十亿美元的生产力损失成本是由于声乐疾病。这个项目的目标是评估未经测试的建模假设的有效性，经常采用。 将在一个7.5倍于真人大小的声带可控动态模型中进行流体流动研究。 声门内的粘性流动现象将通过使用粒子图像测速（PIV）和激光多普勒测速（LDV）的空间和时间流场数据进行研究。 将进行参数结构分析，以确定声带结构对8个几何参数和6个组织参数的敏感性，其中有几个参数从未测量过。 从这项研究中获得的知识将提高未来模型的准确性和有效性，最终导致创建临床相关的人类语音模型。 除了演讲制作，研究结果将与流体-结构相互作用和生物医学工程等更广泛的领域相关。普渡大学现有的推广计划将确保除了工程和社会影响外，研究将对代表性不足的群体的参与，本科生的研究参与和K-12推广计划产生更广泛的影响。 普渡大学的夏季本科生研究奖学金（SURF）计划为本科生提供有意义的指导研究，普渡大学的少数民族工程组织特别鼓励女性和少数民族的参与。 参与工程推广计划，如“想象站”，“中学思想”，和普渡大学的EPICS将促进K-12学生接触工程的兴奋在年轻的时候。 与EPICS一起创建的人类语音过程的功能模型将用于本科语音语言和听力研究（SLHS）课程的教学。