WELT: Using Graphics Generation in Linguistic Fieldwork
Morgan Ulinski, Anusha Balakrishnan, Bob Coyne, Julia Hirschberg, Owen Rambow · 2014
We describe the WordsEye Linguistics tool (WELT), a novel tool for the docu-mentation and preservation of endangered languages. WELT is based on Words-Eye (Coyne and Sproat, 2001), a text-to-scene tool that automatically generates 3D scenes from written input. WELT has two modes of operation. In the first mode, En-glish input automatically generates a pic-ture which can be used to elicit a de-scription in the target language. In the second mode, the linguist formally docu-ments the grammar of an endangered lan-guage, thereby creating a system that takes input in the endangered language and gen-erates a picture according to the grammar; the picture can then be used to verify the grammar with native speakers. We will demonstrate WELT’s use on scenarios in-volving Arrernte and Nahuatl. 1