Reflections on software and technology for language documentation

Abstract

Technological developments in the last decades enabled an unprecedented growth in volumes and quality of collected language data. Emerging challenges include ensuring the longevity of the records, making them accessible and reusable for fellow researchers as well as for the speech communities. These records are robust research data on which verifiable claims can be based and on which future research can be built, and are the basis for revitalization of cultural practices, including language and music performance. Recording, storage and analysis technologies become more lightweight and portable, allowing language speakers to actively participate in documentation activities. This also results in growing needs for training and support, and thus more interaction and collaboration between linguists, developers and speakers. Both cutting-edge speech technologies and crowdsourcing methods can be effectively used to overcome bottlenecks between different stages of analysis. While the endeavour to develop a single all-purpose integrated workbench for documentary linguists may not be achievable, investing in robust open interchange formats that can be accessed and enriched by independent pieces of software seems more promising for the near future.