This article describes IReadWeb system, which is based on existing WebAnyWhere technology. The existing WebAnyWhere system uses depth-first search (DFS) to traverse the Document Object Model (DOM) during the Web surfing task. DFS uses an exhaustive search and crawls through an entire page until it identifies the target node thereby greatly increasing the response time to users. We developed a user-experienced based algorithm, which, unlike DFS, exploits pre-fetched information stored in a local cache to speed up the browsing task. The performance of IReadWeb is thoroughly evaluated and compared against WebAnyWhere by using a sizeable sample of blind native Arabic speakers. The experimental results show that IReadWeb outperformed WebAnyWhere in attaining fast response speed.

Text-to-speech synthesis is the process of converting written text to speech. The lack of research on the growth of and the need for the Arabic language is notable. Therefore, this paper reports an empirical study that systematically compares two screen readers, namely, NonVisual Desktop Access (NVDA) and IBSAR. We measured the quality of these two systems in terms of standard pronunciation and intelligibility tests with visually impaired or blind people. The results revealed that NVDA outperformed IBSAR on the pronunciation tests. However, both systems gave competitive performance on the intelligibility tests.