The main focus of the ISLE Meta Data Initiative (IMDI) is the metadata annotation of multi‐media and multi‐modal language resources, and written language corpora. This metadata set is the results of the ISLE project (International Standard for Language Engineering) and is one of the most detailed and universal proposals for a set of metadata descriptions for linguistic resources.

The IMDI has several sets of metadata, for example: Catalogue metadata and Session metadata, which describe different levels of information. The Catalogue metadata set describes more the abstract level of the resources. It contains various metadata fields, such as title, authors, publisher, language etc, to range linguistic resources. Moreover, it helps to find some particular resources in an archive by means of the fields.

The Session metadata set describes individual recording of corpus components, for example administrative information, information about circumstances and conditions of the linguistic event, groups of resources, information about the media files, personal identification information (e.g. who was involved in the session), etc. Both of the metadata sets are formalized by the IMDI XML Schema.

The IMDI metadatas set are particularly suitable for annotating multi‐modal corpora. It is very comprehensive and conceptual, and includes other widely known metadata sets like Dublin Core or OLAC. Arbil is an open source tool which can be used for browsing, editing, and creating new metadata according to the IMDI metadata schema.