Leveraging Digital Library Infrastructure to Build a Language Archive

Building a digital language archive requires a number of steps to ensure collecting, describing, preserving, and providing access to language data in effective and efficient ways. The Computational Resource for South Asian Languages (CoRSAL) group has partnered with the University of North Texas (UNT) Digital Library to build a series of interconnected digital collections that leverage existing UNT technical and metadata infrastructure to provide access to data from and for various language communities. This article introduces the reader to the background of this project and discusses some of the important for representing language materials areas where UNT metadata has needed flexibility to better fit the needs of intended audiences. These areas include a workflow for standardized language representation (the Language field), defining roles for persons related to the item (Creator and Contributor fields), and representing interconnections between related items (the Relation field). Although further work is needed to improve language data representation in the CoRSAL digital language archive, we believe the model adopted by our team and lessons learned could benefit others in the language archiving community.