Designing assistive technologies for medication regimes in care settings

This paper presents some early design work of the Care in the Digital Community research project begun under the EPSRC IRC Network project Equator. Gaining a comprehensive understanding of user requirements in care settings poses interesting methodological challenges. This paper details some methodological options for working in the domestic domain and documents the translation of research into design recommendations. We report on the importance of medication issues in a hostel for former psychiatric patients and present an early prototype of a medication manager designed to be sensitive to the particular requirements of the setting.     

语音处理技术在语言学习中的应用

语音处理技术应用于语言的学习,可以改进、加强和丰富传统的语言学习方法,有利于提高学习效率;然而,如何有效地利用这些技术,建立各种计算机辅助语言学习 ＣＡＬＬ（Ｃｏｍｐｕｔｅｒ－Ａｉｄｅｄ Ｌａｎｇｕａｇｅ　Ｌｅａｒｎｉｎｇ）系统,是当前语音处理技术研究发展的一个重要方向。该文通过对传统的语言教学方法中所涉及到的主要因素的分析,研究了语音处理技术在语言学习系统中应用的主要层次和方法,以及使其更有效等方面的问题。     

蒙古语语音识别系统语言模型的设计

目前,蒙古语语音识别的研究尚处于空白阶段,因此蒙古语语音识别系统的研究与开发具有重要意义。而语言模型的确立是语音识别系统中最重要的环节之一。本文根据自己的实践,通过实验的方法最终确立了蒙古语、大量词汇语音识别系统中适宜的语言模型。     

A commercial perspective on universal access and assistive technology: towards implementation

This paper presents one companys perspective on the implementation and provision of universal access (UA) and assistive technology in an industrial setting. The paper addresses the need to provide accessible work-places and also accessible customer services, from legal, commercial and ethical standpoints. The company in question, Royal Mail, is one of the UKs largest employers and service providers and so has been able to gather employee and customer data often unavailable to smaller organisations.     

Involving domain experts in assistive technology research

Teams engaging in assistive technology research should include expertise in the domain of disability itself, in addition to other areas of expertise that are more typical in human–computer interaction (HCI) research, such as computer science and psychology. However, unexpected problems can arise when HCI researchers do not adequately plan the involvement of domain experts in a research project. Although many research teams have included domain experts when designing assistive technologies, there has been little work published on how to best involve these experts in the research process. This paper is a first step towards filling that void. Based on the authors’ own experiences involving domain experts in research, as well as those documented in the literature, five types of domain experts and three broad roles that domain experts can play are identified, and five guidelines for their involvement are presented. This analysis will be useful to anyone in the assistive technology and universal accessibility communities, especially those who are in the early stages of conducting research in this area. It is intended to lay the foundation of best practices for involving domain experts in assistive technology research.     

基于C++ Builder的田野语音采集系统的设计与实现

本文结合我国少数民族语言濒危现象,以C++ Builder 6.0为开发工具,设计开发了田野语音采集系统,用于保护语言文化遗产。该系统主要功能包括:数据管理、音频设置与分析、快速采集、普通采集、系统管理、数据库的备份与恢复等。相对其他方言采集工具,该系统能有效提高语音采集的质量和效率,便于对语音数据库进行科学管理,具有一定的应用价值。     

Generating American Sign Language animation: overcoming misconceptions and technical challenges

Misconceptions about the English literacy rates of deaf Americans, the linguistic structure of American Sign Language (ASL), and the suitability of traditional machine translation (MT) technology to ASL have slowed the development of English-to-ASL MT systems for use in accessibility applications. This article traces the progress of a new English-to-ASL MT project targeted to translating texts important for literacy and user-interface applications. These texts include ASL phenomena called “classifier predicates.” Challenges in producing classifier predicates, novel solutions to these challenges, and applications of this technology to the design of user-interfaces accessible to deaf users will be discussed.     

一种精确检测语音端点的方法

端点检测是语音识别中的一项关键技术,端点检测的准确性对语音识别的性能有很大影响,特别是对端点检测比较敏感的语音识别算法。本文引用窗长动态变化的端点检测技术,并将传统的双门限端点检测算法和窗长动态改变的端点检测技术结合起来用于语音端点检测。大量实验表明这种技术可以比较精确的检测语音端点,特别是地检测语音的起始端点中有很大的优势。使用改进后的语音端点检测技术,可以有效地提高语音识别率。     

Mechanisms for actuated assistive hip orthoses

Mobility is often a central problem for people having muscle weaknesses. The need for new devices to assist walking and walk related activities is therefore growing. Lower limb actuated orthoses have already proven their positive impact with paraplegic patients and are potentially promising for assisting people with weak muscles. However, the transfer from the existing systems of mobilization towards assistance implies several technical challenges as the seamless integration and the reduction of power consumption. In this paper two assistive orthoses which use different types of actuation mechanisms are presented and discussed. The first one is based on a ball screw and an excavator-like mechanism while the second one is based on a double differential actuation. Their technical capabilities are compared and contextualized for diverse activities. Objective characteristics such as the range of motion of the devices, the transparency, the maximal torque that they can provide or the RMS torque during cyclic trajectories are compared to point out which device is better adapted for specific situations.     

汉语文语转换及语音编解码的基于DSP的综合实时处理

提出了汉语文语转换和语音编解码的基于DSP的实时综合处理的思想。介绍了实时综合处理的基于DSP的硬件与软件实现。文章介绍的基于DSP的实时综合处理具有以下几个方面的特点：(1)在单片DSF和不超过IMB的DSP最小系统上实现了一个任意文本的汉语文语转换系统;(2)为提高语音合成的自然度,提供了灵活的语音定制功能;(3)根据通信信道容最大小,实现可灵活选择的采甩语音合成方式的语音通信和语音编码方式的语音通信;(4)基于DSP的实时综合处理的在线重构;(5)基于RS232的通用异步串口的包括命令解释引擎、合成引擎、编码引擎等的模块化DSP软件结构。     

电话语音识别系统

讨论了一个典型的电话语音识别系统的设计与实现。首先提出了系统的整体模型,之后给出了系统的整体结构和系统的主要模块。较详细地介绍了声学层识别的速度优化和电话信道补偿问题,以及语言模型的容错算法和基于语义依存关系的容错理解模型,并对系统各层次模块性能进行了实验测试。     

一种数字语音处理研究平台的设计

设计了一种支持多视图和多种数据接口的研究平台。该平台采用了模块组合式的体系结构，使得新的语音处理算法能够便捷地加入到平台中。所有模块被分为内部模块和外部模块。内部模块集成语音的数据接口和可视化功能，外部模块则实现语音的分析功能。该平台采用了纵版的显示方式，不同算法得到的结果被垂直排列显示在一个窗口内，有利于对比分析，还内建了支持基于XML的语音标注格式，能够被直接应用到语料库建设和语音分析等领域。     

Algebraic tools for language processing

A language space provides a unified framework to deal with the properties of language constructs by associating them with their specification rules. The concrete syntax is represented by segments of the language space. The semantics is given by derived operations of the algebras where these constructs are interpreted by the processing tools operating on the language space. We examine in this paper only processing tools that collect syntactic information over the language space. Tools involved in semantics processing such as translators and interpreters are also integrated in the language space but are not discussed here.     

信息技术与语文教学整合的研究

信息技术与语文教学的整合是真正意义上的语文教学革命,它改变了教学信息的传递,使教学信息的线性传递方式转变为非线性传递方式。用多种媒体形式的信息组织语文教学,使得教学效果更加直观生动,从而与以往的语文教学有很大的不同。信息技术与语文教学的整合强调寓教于乐,它具有媒体的多样性、交互性、集成性能更好地培养学生自主探索的能力,学生的语文素养提高了,听、说、读、写等语文教学内容在信息技术的支持下获得了新的发展。     

基于MS Agent的语音识别与合成

语音合成与语音识别提供了全新的人机交互方式,在信息处理、教育与商务应用等方面已显示出了巨大的优势。本文分析了MS Agent的工作原理,论述了基于MS Agent的语音系统的开发步骤,给出了系统运行示意图。     

A multimode navigation system for an assistive robotics project

Assistive technology is an emerging area, where robotic devices can help individuals with motor disabilities to achieve independence in daily activities. This paper deals with a system that provides remote control of Sony AIBO, a commercial mobile robot, within the assistive project ASPICE. The robot can be controlled by various input devices, including a Brain-Computer Interface. AIBO has been chosen for its friendly-looking aspect, in order to ease interaction with the patients. The development of the project is described by focusing on the design of the robot navigation system. Single step, semi-autonomous and autonomous navigation modes have been realized to provide different levels of control. Automatic collision avoidance is integrated in all cases. Other features of the system, such as the video feedback from the robotic platform to the user, and the use of AIBO as communication aid, are briefly described. The performance of the navigation system is shown by simulations as well as experiments. The system has been clinically validated, in order to obtain a definitive assessment through patient feedback.

一个面向语音识别的云南民族口音普通话语音数据库

介绍了一个以语音识别为目的的云南民族口音普通话语音数据库。当前,语音识别技术要走向实用必须解决用户情况多样性带来的鲁棒性问题,通常把这个问题简要地归结为“男女老幼”和“南腔北调”。作为民族文化大省的云南,共有25个少数民族,广大少数民族同胞在说普通话时明显带有地方民族口音,云南民族口音普通话语音识别研究是用户情况多样性研究的重要内容,而为之建立云南民族口音普通话语音数据库是该研究的重要基础和先决条件。     

用于车载无线终端的嵌入式语音处理系统

介绍了一个用于汽车环境的无线终端中利用语音技术进行语音识别拨号、语音合成和语音提示的系统.系统包括两个主要的模块:语音处理模块和蓝牙通信模块.其中蓝牙通信模块的功能是与具有蓝牙接口的手机进行通信,包括连接手机进行通话,下载手机内的电话号码本并传送给语音处理模块;语音处理模块完成语音识别、语音合成、语音提示、利用语音压缩编解码进行通话录放以及号码查询等功能,并控制整个系统的流程.该系统可以实现对手机电活本的下载并在线生成识别词表,识别词表容量可达1000词;在600词情况下的实验结果表明系统的识别率大于97%;系统基于SoC架构,具有高集成度和高稳定性的特点.