For most of the world’s languages, no primary data are available, even as many languages are disappearing. Throughout the last two decades, however, language documentation projects have produced substantial amounts of primary data from a wide variety of endangered languages. These resources are still in the early days of their exploration. One of the factors that makes them hard to use is a relative lack of standardized annotation conventions. In this paper, we will describe common practices in existing corpora in order to facilitate their future processing. After a brief introduction of the main formats used for annotation files, we will focus on commonly used tiers in the widespread ELAN and Toolbox formats. Minimally, corpora from language documentation contain a transcription tier and an aligned translation tier, which means they constitute parallel corpora. Additional common annotations include named references, morpheme separation, morpheme-by-morpheme glosses, part-of-speech tags and notes.

This paper describes a collection of 20k ELAN annotation files harvested from five different endangered language archives. The ELAN files form a very heterogeneous set, but the hierarchical configuration of their tiers allow, in conjunction with the tier content, to identify transcriptions, translations, and glosses. These transcriptions, translations, and glosses are queryable across archives. Small analyses of graphemes (transcription tier), grammatical and lexical glosses (gloss tier), and semantic concepts (translation tier) show the viability of the approach. The use of identifiers from OLAC, Wikidata and Glottolog allows for a better integration of the data from these archives into the Linguistic Linked Open Data Cloud.