A prototype: Reading of Web documents for visually impaired persons

Describes a system for knowledge-based information extraction from Web-documents, reading HTML and XML structured documents using logic inference and text categorization. The system uses a multimedia speech synthesizer, in the windows environment. The users could be normal persons searching filtered information without paying their whole attention, display-disabled readers or visually impaired persons. In the first stage, structured digital texts like public domain books (e-books) are read and translated to voice; their headings, chapter names, subdivisions, keywords, and paragraphs are read. The user personalizes his filters using keywords and the user can move inside the text through the paragraphs while listening to the contents. This project mainly increases the availability of electronic-texts for persons with a reading handicap.