Acoustic Mapping of Visual Text Signals through Advanced Text-to- Speech: the Case of Font Size

Current Text-to-Speech systems, commonly used for document accessibility, do not include an effective and standard acoustic provision of the visual typographic cues embedded in them. In this work, we first introduce the text signals (i.e. the writing devices that emphasize aspects of a text’s content or structure) along with an appropriate architecture for the structure of documents as well as the main principles and technological topics of document accessibility. Then, the emerging technological approach of Document-toAudio (DtA) we have developed is presented. DtA essentially constitutes the next generation of Text-to-Speech systems that supports the acoustic mapping of visual text-signals and thus provides much better accessibility to documents. Finally, for the case of the font size in the typographic layer, we present the results of two quantitative approaches (direct mapping and emotional-based mapping) for the rendering of text signals through DtA. Key-Words: Text-to-Speech, Text Signals, Document-to-Audio, Document Visual Elements, Accessibility