Assessing Agreement Level Between Forced Alignment Models with Data from Endangered Language Documentation Corpora
The authors used Yoloxochitl Mixtec elicitation material developed as part of Amith's National Science Foundation grant to test the ability of automated tools for phonetic segmentation to match hand-labeled segmentation. The success achieved by the automated alignment suggests that even the relative small sets of materials produced by language documentation efforts may be effectively labeled through automated processes.
DiCanio, Christian, Jonathon D. Amith, H. Timothy Bunnell, Rey Castillo Garcia, Hosung Nam, and Douglas H. Whalen. “Assessing Agreement Level Between Forced Alignment Models with Data from Endangered Language Documentation Corpora.” Proceedings of InterSpeech. Portland: 1-4. 2012.