Content-Driven Techniques for Non-Visual Web Access

万维网已经发展成为信息传播、娱乐、商业和教育不可或缺的媒介。然而，大多数浏览软件的图形特性，加上网络内容的多样性和复杂性，限制了整个视觉障碍群体使用这项技术的机会。现有的基于文本到语音转换的音频浏览器（例如，屏幕阅读器）不能描述文档内容的概念组织，也不能让用户选择文档的一部分来收听。因此，有视觉障碍的人很难执行常见的任务，比如区分主题或关联类似的项目，而这些是理解文档组织的关键，因此他们会浪费大量的时间和精力去听不相关的信息。在这个项目中，PI和他的团队将开发、测试和推广HearSay网络浏览器，这将使视力障碍者的浏览体验更接近正常人。HearSay基于自动化技术，将web文档的内容结构化为由逻辑相关项组成的标记分区。通过启用交互式语音驱动的引导探索，系统显示文档的标记内容，用户选择要听的内容部分以及何时导航到新页面，HearSay将大大减少非视觉浏览的麻烦。此外，对于重复的浏览任务，HearSay将允许用户以不同的方式创建和检索个性化的内容，从基于内容的语音标记页面中选定的分区到强大的个人信息助手，根据用户的命令收集和呈现用户定义的信息。为了确保HearSay系统真的像宣传的那样对目标群体有用，PIs与纽约州亨普斯特德的海伦·凯勒盲人服务机构建立了合作关系，该机构为视力障碍人士提供培训，并将为HearSay的设计和评估提供咨询。更广泛的影响：这项研究将产生新的算法和强大的技术，使最终用户能够使用音频浏览网络内容。这一功能对视力受损的人来说尤其有价值，因为它使他们能够自己浏览和定制web内容，但它对小型设备的移动用户也很有用，减少了他们对专业内容提供商的依赖。这些项目将为Blackboard教育系统开发一个特别版本的HearSay，并向教育机构进行宣传和分发，以改善视障学生获取教育材料的机会。与这个项目协调，将组织两个关于无障碍技术的讲习班，一个侧重于一般无障碍技术，另一个侧重于研究生和成人教育无障碍技术。这些讲习班将有助于提高人们对视力障碍者面临的问题的认识，以及基于内容的技术在改善移动和/或残疾用户获取信息方面的潜力。

The World Wide Web has evolved into an indispensable medium for dissemination of information, entertainment, commerce and education. However, the graphical nature of most browsing software, coupled with the diversity and complexity of web content, have limited access to this technology for an entire community of people with visual disabilities. Existing audio browsers that are based on text-to-speech conversion (e.g., screen readers) are not capable of describing the conceptual organization of a document's content or of letting a user select parts of a document to listen to. As a result, people with visual disabilities can find it difficult to perform common tasks, such as distinguishing topics or correlating similar items, which are key to understanding the organization of documents, and so waste considerable time and attention listening to irrelevant information. In this project the PI and his team will develop, test and disseminate the HearSay web browser, which will bring the browsing experience of people with visual disabilities closer to that of sighted people. HearSay is based on automated techniques for structuring the content of web documents into labeled partitions consisting of logically related items. By enabling interactive speech-driven guided exploration in which the system presents the document's labeled content and the user selects which parts of the content to listen to and when to navigate to a new page, HearSay will make non-visual browsing far less cumbersome. Furthermore, for repetitive browsing tasks, HearSay will let users create and retrieve personalized content in different ways, ranging from content-based voice-marking of selected partitions in a page to powerful personal information assistants that gather and present user-defined information at the user's command. To assure that the HearSay system is really useful "as advertised" to the intended community, the PIs have established a collaboration with Helen Keller Services for the Blind in Hempstead, NY, which trains people with visual disabilities, and will consult on the design and evaluation of HearSay. Broader Impacts: This research will result in new algorithms and powerful technologies enabling end users to navigate web content using audio. This capability will be especially valuable to people who are visually impaired in that it will enable them to browse and customize web content by themselves, but it will also be useful for mobile users of small-form devices, reducing their dependence on specialized content providers. The PIs will develop a special version of HearSay for use with the Blackboard educational system, which will be advertised and disseminated to educational institutions so as to improve access to educational materials for students with visual disabilities. In coordination with this project, two workshops on technology for accessibility will be organized, one focusing on accessibility technology in general and the second on accessibility technology for postgraduate and adult education. These workshops will serve to raise awareness of the issues faced by those with visual disabilities and the potential for content-based techniques to improve information access for users who are mobile and/or have disabilities.