Capturing variation in a language documentation corpus

Variation in language documentation corpora is pervasive, but usually little reflected in annotations or scholarly writing. In this talk, I illustrate different levels of variation—phonetic, lexical, and conceptual—in my corpus of Gyeli, and endangered Bantu language of Cameroon. I further suggest how to code different types of variation in ELAN. (session 1.4.4) 44777.mp3