Sharing community interpreting corpora

This article presents results of a project which looks into new ways of publishing and sharing community interpreting corpora. Section 2 introduces CoPilot, a pilot database containing material from two corpora of interpreted doctor-­patient communications and one corpus of interpreted court proceedings. Section 3 discusses the technical challenges involved in making such corpora from different sources available in a common environment. In Section 4, we sketch some of the properties such an environment should have in order to be usable and useful for the community. Section 5, finally, uses the annotation of the language of an utterance and the annotation of translation status as two examples illustrating the benefits and problems involved in annotating community interpreting corpora across language and corpus boundaries.