Ensuring accessibility to scientific material has always been a challenging task and can be considered a major obstacle for full inclusiveness in education in the fields of science, technology, engineering and mathematics (the traditional STEM subjects), in particular at the late secondary and tertiary stage. Traditionally mathematical content was made accessible either via audio books recorded by subject specialists, or by manually generating different print-like formats, e.g., large print, Braille, high contrast, etc. As teaching moves more and more towards the provision of online material and courses, the traditional means of making content accessible are all but obsolete. To avoid the risk that modern technology might create an even higher obstacle for inclusive education, it is important to ensure accessibility of scientific web content without the need for expensive, specialist software.
Dr Volker's research concentrates on making mathematical formulae accessible for visually and print impaired readers, which is part of a larger body of work on providing assistive technology solutions for STEM content. Since formulas can be generated in many ways, from a variety of markup languages (LaTeX, AsciiMath, MathML, etc.), and formulas with the same visual appearance can be constructed quite differently, an important pre-requisite for generating homogeneous and meaningful presentations in alternative formats such as speech, Braille or as interactive applications, is a semantic analysis and reconstruction of the presentation form. Hence at the core of his work is the definition and generation of a semantically rich representation by means of pattern recognition techniques and domain-specifically selected heuristics. This serves as a basis for innovative creation of alternative formats. Since modern web technology enables embedding of sufficient hidden semantic information, content can not only be made accessible in a traditional way, but allows for a manifold of personalisable views that can support learners with diverse special needs. All these ideas are implemented in a system Speech Rule Engine (SRE) and practically tested in the MathJax library for rendering mathematics in web browsers.