Smartphone-Assisted Adaptive Speech Enhancement and Auditory Training of Hearing

智能手机辅助自适应语音增强和听力训练

• 批准号: 8884694
• 负责人: Issa Panahi
• 金额: $ 26.12万
• 依托单位: UNIVERSITY OF TEXAS DALLAS
• 依托单位国家: 美国
• 财政年份: 2014
• 资助国家: 美国
• 起止时间: 2014-06-01 至 2016-05-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/8884694
• 关键词: Address; Age; Algorithms; Auditory; Cellular Phone; Characteristics; Classification; Clinical Data; Communication; Complex; Crowding; Data; Development; Devices; Educational process of instructing; Emergency Situation; Environment; Environmental Wind; Food; Frequencies; Future; Goals; Health; Hearing; Hearing Aids; Hearing problem; Individual; Measures; Memory; Methods; Noise; Outcome; Outcome Study; Perception; Performance; Power Sources; Process; Protocols documentation; Reporting; Research; Research Activity; Residual state; Resolution; Restaurants; Running; Signal Transduction; Solutions; Source; Speech; Speech Intelligibility; Technology; Testing; Training; Training Programs; Voice; Wireless Technology; Work; base; computing resources; design; flexibility; hearing impairment; improved; innovation; instrument; programs; signal processing; simulation; sound; trafficking

DESCRIPTION (provided by applicant): Existing hearing aid devices (HAD) with built-in processors and features perform multiple signal processing tasks. However, due to design constraints such as device size, computational capability of processors, memory size, and number of microphones, application of more sophisticated signal processing algorithms that could improve performance of HAD are either oversimplified or ignored. To circumvent these design constraints, we propose the use of smartphone and wireless technologies to assist and improve performance of HAD. The thrust of this proposal is to provide an assistive processing paradigm for enhancing the quality and intelligibility of speech for an individual HAD user in various noisy environments. The proposed research activity takes a two-pronged approach under one overarching principle: improvement of performance of HAD for individual user by utilizing the capabilities of smartphone and wireless technologies. The first component of research provides an alternative approach where the signal processing tasks of the HAD processors are assisted and enhanced by the adaptive algorithms run on the processor of a smartphone. The proposed approach will have the advantage of tapping into the computing resources, features, and microphones of smartphone for running complex signal processing algorithms without placing heavy computation and memory burdens on the signal processors of HAD. When the smartphone is not in use, the HAD operates alone running the previously saved-in algorithms. The second component of research is complementary to the first research component. It is on specific application of interactive auditory training methods for the individua user of HAD. The auditory training algorithm resides on the smartphone and is available to the user as a new application. By running this application at the user discretion, the HAD user can train his hearing ability for better perception and recognition of spoken words in the speech that might be corrupted by various types of residual or interrupting short-duration noise signals. An innovation of the proposed study lies in that while the first research component improves enhancement of quality and intelligibility of speech simultaneously, the second research component individualizes the HAD through user controls. The development of interactive auditory training protocols which can be operated on a smartphone is innovative because it takes advantage of commonly used technology. Therefore, the users are more receptive to engage in the training programs because it is readily available and can be done in brief moments, frequently throughout the day. For example, some may even engage in training while waiting for a bus or while in line to order food. The proposed research offers the users a personalized and individually tunable application of HAD with improved performance. Furthermore, application of smartphone working wirelessly with HAD offers a flexible platform for educational purposes, teaching, and future research innovations. The proposed study and its outcomes will pave the way for many new research ideas and useful development to help people of different age with hearing disorders in future.

描述（由申请人提供）：具有内置处理器和功能的现有助听器设备（HAD）执行多个信号处理任务。然而，由于诸如设备尺寸、处理器的计算能力、存储器大小和麦克风的数量等设计约束，可以提高HAD性能的更复杂的信号处理算法的应用被过度简化或忽略。为了规避这些设计限制，我们建议使用智能手机和无线技术，以协助和提高性能的HAD。这项建议的主旨是提供一个辅助处理范例，以提高个别HAD用户在各种嘈杂环境中的语音质量和清晰度。拟议的研究活动在一个总体原则下采取双管齐下的方法：通过利用智能手机和无线技术的功能，提高个人用户的HAD性能。研究的第一个组成部分提供了一种替代方法，其中HAD处理器的信号处理任务由智能手机处理器上运行的自适应算法辅助和增强。所提出的方法将具有利用智能手机的计算资源、功能和麦克风来运行复杂信号处理算法的优点，而不会对HAD的信号处理器造成沉重的计算和存储负担。当智能手机不使用时，HAD单独运行先前保存的算法。第二个研究部分是对第一个研究部分的补充。这是交互式听觉训练方法在HAD个人用户中的具体应用。听觉训练算法驻留在智能手机上，并作为新应用程序提供给用户。通过在用户的判断下运行此应用程序，HAD用户可以训练他的听力能力，以便更好地感知和识别语音中可能被各种类型的残留或中断的短持续时间噪声信号破坏的口语单词。所提出的研究的创新之处在于，虽然第一个研究组件提高了语音的质量和清晰度的同时增强，第二个研究组件个性化的HAD通过用户控制。可以在智能手机上操作的交互式听觉训练协议的开发是创新的，因为它利用了常用的技术。因此，用户更容易接受参与培训计划，因为它是现成的，可以在短时间内完成，经常在一天中。例如，有些人甚至可能在等公共汽车或排队点菜时参加培训。建议的研究为用户提供了一个个性化的和单独可调的应用程序的HAD与改进的性能。此外，智能手机与HAD无线工作的应用为教育目的，教学和未来的研究创新提供了一个灵活的平台。这项拟议的研究及其结果将为许多新的研究思路和有用的发展铺平道路，以帮助未来不同年龄的听力障碍患者。