READING REMEDIATION USING COMPUTER SPEECH RECOGNITION

使用计算机语音识别进行阅读纠正

• 批准号: 6143155
• 负责人: DONALD G STEELY
• 金额: $ 9.99万
• 依托单位: OREGON CENTER FOR APPLIED SCIENCE, INC.
• 依托单位国家: 美国
• 财政年份: 2000
• 资助国家: 美国
• 起止时间: 2000-08-01 至 2001-07-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/6143155
• 关键词: computer assisted instruction; computer program /software; computer system design /evaluation; educational resource design /development; human subject; reading; reading disorder; remedial /special education; speech recognition; training level

Reading failure occurs in our schools at unacceptably high levels, causing serious and widespread social, political, and economic problems. This occurs despite the availability of effective instructional programs. Unfortunately, teachers often do not have the necessary training or time to implement these programs. In addition, it is sometimes not cost effective to use these programs when student groups are small or scattered. This project will incorporate advanced speech recognition technology with the computer delivery of effective research-proven instruction to effectively and economically teach basic reading skills to learning disabled readers. In Phase I, ten lessons will be produced and evaluated to demonstrate the feasibility of this approach. These lessons will encompass a wide range of decoding skill applications needed by learning disabled readers in grades one through five. The prototype feasibility will be evaluated with 80 at-risk readers in grade two. This prototype will be a stand-alone, turn-key instructional system, including speech recognition software, sound card, computer, monitor, and microphone. In Phase II, we will develop a 150-lesson remedial reading program and test its efficacy in a randomized control trial. PROPOSED COMMERCIAL APPLICATIONS: This project will lead to the production of a computer-based program that provides a comprehensive reading program for learning disabled and remedial readers. It will be attractive to schools, tutorial centers, and home markets.

我们学校的阅读失败率高得令人无法接受，造成了严重而广泛的社会、政治和经济问题。尽管存在有效的教学计划，但这种情况还是发生了。不幸的是，教师往往没有必要的培训或时间来实施这些计划。此外，当学生群体较小或分散时，使用这些程序有时并不划算。该项目将先进的语音识别技术与计算机提供经过研究验证的有效指令相结合，以有效且经济地向学习障碍读者传授基本阅读技能。在第一阶段，将制作并评估十个课程，以证明这种方法的可行性。这些课程将涵盖一年级至五年级学习残疾读者所需的广泛解码技能应用。原型的可行性将由 80 名二年级高危读者进行评估。该原型将是一个独立的交钥匙教学系统，包括语音识别软件、声卡、计算机、显示器和麦克风。在第二阶段，我们将开发一个 150 课时的辅导阅读计划，并在随机对照试验中测试其效果。拟议的商业应用：该项目将产生一个基于计算机的程序，为学习障碍和补习读者提供全面的阅读程序。它将对学校、辅导中心和家庭市场有吸引力。