NLP for Related Languages

Explore the potential of leveraging language relatedness to bridge the technological gap between high- and low-resource languages in this insightful talk by Nate Robinson. Delve into the multidimensional aspects of language relationships, examining how phonology, morphosyntax, acoustics, and phylogenetics can be utilized to enhance natural language processing technologies for under-resourced languages. Discover findings on the effectiveness of morphosyntactic relatedness in machine translation, the role of phonological information in language processing applications, and the benefits of cross-lingual data augmentation for low-resource speech technologies. Gain insights into how algorithms modeling linguistic change can contribute to computational linguistics research. Learn about the implications of these findings for groups of related low-resource languages, including Creole languages of the African diaspora, Arabic language varieties, and Sinitic languages. Presented by a mathematician-turned-researcher at CMU's Language Technologies Institute, this talk offers a creative mathematical approach to addressing linguistic challenges and developing tools that break down language barriers for speakers of under-represented languages.

Nate Robinson - “NLP for Related Languages”