Computer-based auditory skill building program for aural (re)habilitation

描述(由申请人提供)：在过去的十年里，两个进步使得听力损失患者获得听力技能和可理解的语音变得更容易实现。第一种是普遍的新生儿听力筛查，第二种是人工耳蜗术。在18个月前接受CI检查的儿童中，有90%的人语言清晰。然而，对于这些接受CI的儿童和其他接受CI的人来说，强化的听力(康复)训练至关重要，才能充分认识到人工耳蜗术的好处。与言语和语言病理学家面对面的培训既昂贵又耗时，而且可能并不总是可用的，因为客户可能生活在没有专家的农村地区，或者儿童可能被安置在没有内部AR专家的公立学校。市面上有几种可以自我管理的听觉(康复)工具。其中许多都是专门为成人演讲者设计的。俄勒冈健康与科学大学(Oregon Health&Science University)口语理解中心(CSLU)的研究人员和塔克-马克森口语学校(Tucker-Maxon Oral School)的专家开发了一种可供儿童使用的工具--听力技能辅导。在第一阶段，我们将大大扩展这一听力技能导师(LST)的能力，并对其表现进行评估。LST最重要的改进是，它将包含针对各种音素听力目标的适应性计算机指导课程，允许听力受损的个人在没有专家广泛指导的情况下获得听力技能。此外，新系统为音素收听目标提供了改进的语音(无论是自然语音还是改进的合成语音)；除了或代替单词替代使用图像替代；使用自然说话的脸，而不是合成的说话脸。在这个第一阶段-STTR项目中，我们将重点放在语音识别和识别技能的培养上。在第二阶段，我们将扩大LST，将超段目标和全球理解活动包括在内。超音段目标包括基音和基音轮廓、幅度和语速。整体理解包括理解对话中的单词，并将在安静和不利的听力条件下呈现，例如在噪音中或在竞争对话中(喋喋不休)。此外，在第二阶段，我们将建立一个以计算机为基础的评估工具，以确定用户需要学习的具体目标。评估工具将使用自上而下的测试方法(或阶梯法)，根据选定的刺激数量确定用户的入门水平。在美国，大约22,000名成年人和近15,000名儿童接受了人工耳蜗术，这些人需要强化听觉(恢复)训练，以充分实现人工耳蜗术的潜在好处，并获得清晰的语音。与专家面对面的AR培训费用高昂，并不总是由保险覆盖，而且并不总是可行，因为客户可能生活在没有专家的农村地区，或者儿童可能被安置在没有内部AR专家的公立学校。因此，一个计算机引导的程序，允许在没有专家帮助的情况下进行听觉训练是必要的。拟议的项目将创建和评估这样一个项目，重点是学习音素辨别和识别。它使用自然和高质量的合成声音，用适合儿童的图片显示可选的备选方案，并具有自然的对话脸选项。

DESCRIPTION (provided by applicant): In the past decade, two advances have made the acquisition of listening skills and intelligible speech by individuals with hearing loss more attainable. The first is universal newborn hearing screening and the second is cochlear implantation (CI). 90% of children who receive a CI before they are 18 months old attain intelligible speech. However, intensive aural (re)habilitation training is crucial for these children and other individuals who receive CI's to fully realize the benefit of a cochlear implant. Face-to-face training with a speech and language pathologist is expensive, time-consuming, and may not always be available because clients may live in rural areas where there are no specialists, or children may be placed in public schools without in-house AR specialists. There are several aural (re)habilitation tools available commercially that can be self-administered. Many of these are specifically designed for adult speakers. One tool that is available for children is the Listening Skills Tutor developed by researchers at the Center for Spoken Language Understanding (CSLU) at Oregon Health & Science University and specialists at the Tucker-Maxon Oral School. In Phase I, we will significantly expand the capabilities of this Listening Skills Tutor (LST) and evaluate its performance. The most important improvement of the LST is that it will contain adaptive computer-guided lessons for a wide variety of phoneme listening targets, allowing hearing-impaired individuals to attain listening skills without the extensive guidance of a specialist. In addition, the new system presents phoneme listening targets with improved speech (either natural or improved synthetic voice); uses image alternatives in addition to, or instead of, word alternatives; and uses a natural as opposed to a synthetic talking face. In this Phase I-STTR project we will focus on phonetic discrimination and identification skill building. In Phase II, we will expand the LST to include suprasegmental targets and global comprehension activities. Supra-segmental targets include pitch and pitch contours, amplitude, and speech rate. Global comprehension includes understanding words in conversation and will be presented in quiet and adverse listening conditions such as in noise or with competing conversations (babble). In addition, in Phase II we will build a computer-based assessment tool which will determine what specific targets users need to learn. The assessment tool will use an up-down methodology of testing (or staircase method) which determines the entry level of users based on a selective number of stimuli. PUBLIC HEALTH RELEVANCE In the United States, roughly 22,000 adults and nearly 15,000 children have received cochlear implants and these individuals need intensive auditory (re)habilitation (AR) training to fully realize the potential benefit of a cochlear implant and attain intelligible speech. Face-to-face AR training with a specialist is expensive, not always covered by insurance and not always feasible because clients may live in rural areas where there are no specialists, or children may be placed in public schools without in-house AR specialists. Therefore, a computer-guided program that allows auditory training without the help of a specialist is necessary. The proposed project will create and evaluate such a program, focusing on learning phoneme discrimination and identification. It uses natural and high-quality synthetic voices, displays choice alternatives with child- appropriate pictures, and has a natural talk-face option.