Ontology-Based Annotation of Multimedia Language Data for the Semantic Web

Abstract: There is an increasing interest and effort in preserving and documenting endangered languages. Language data are valuable only when they are well-cataloged, indexed and searchable. Many language data, particularly those of lesser-spoken languages, are collected as audio and video recordings. While multimedia data provide more channels and dimensions to describe a language's function, and gives a better presentation of the cultural system associated with the language of that community, they are not text-based or structured (in binary format), and their semantics is implicit in their content. The content is thus easy for a human being to understand, but difficult for computers to interpret. Hence, there is a great need for a powerful and user-friendly system to annotate multimedia data with text-based, well-structured and searchable metadata. This chapter describes an ontology-based multimedia annotation tool, OntoELAN, that enables annotation of language multimedia data with a linguistic ontology.