Building Mobile Apps for Language Documentation

Abstract

Languages are vanishing from the earth more quickly than linguists can document them. Several factors can account for this language loss, including increasing globalization and the disruption of local language communities, but one reason stands out: the formal documentation of languages has traditionally been the work of trained linguists, who are too few in number to address the current need. However, places with some of the highest numbers of endangered languages, such as Australia and Indonesia, are rapidly adopting mobile technology. In this paper, we develop a tagging feature for collecting and categorizing language data in a recipe-based social media mobile app. We build an infrastructure for relating user-generated language data to multilingual WordNet databases in order to create structured lexicons for low-resource languages.