Digital annotated corpora of Brazilian Indigenous languages with automatic translations (DACILAT)

• Grant number: 22/09158-5
• Support Opportunities: Research Grants - eScience and Data Science Program - Thematic Grants
• Start date: March 01, 2023
• End date: February 29, 2028
• Field of knowledge: Interdisciplinary Subjects
• Principal Investigator: Maria Filomena Spatti Sandalo
• Grantee: Maria Filomena Spatti Sandalo
• Host Institution: Instituto de Estudos da Linguagem (IEL). Universidade Estadual de Campinas (UNICAMP). Campinas , SP, Brazil
• Pesquisadores principais: Charlotte Marie Chambelland Galves
• Associated researchers: Leonel Figueiredo de Alencar Araripe ; Luiz Henrique Lima Veronesi ; Michael Becker ; Pablo Picasso Feliciano de Faria
• Associated scholarship(s): 24/19940-8 - EXPANSION OF THE SET OF SYNTACTICALLY ANNOTATED TEXTS FROM THE TYCHO BRAHE CORPUS - Kadiwéu Language, BP.TT; 24/08126-8 - Consolidation of the Tycho Brahe Platform, BP.TT; 24/00525-0 - Reaching Languages: Documention and Translation of Indigenous Languages, BP.JC; 24/00526-7 - Reaching Languages: Documentation and Translation of Indigenous Languages, BP.JC; 23/16110-1 - ELABORATION OF THE PARSED TEXTS OF THE TYCHO BRAHE CORPUS OF THE NHEENGATU LANGUAGE, BP.TT

Abstract

Drawing on the vast experience of the researchers of the three partner institutions in field work, language documentation, corpus linguistics, computational linguistics, and computer science, our main goal in this project is to build a digital platform to accommodate oral corpora for South American endangered languages. Oral corpora serve as an essential empirical foundation for linguistic research, and they also provide a foundation for literacy projects for the local communities, raising their own meta-linguistic and cultural awareness. Our ultimate goal is to implement automatic translation of Kadiwéu and Nheengatu. Kadiwéu is a language with quite complex morphology, and this is a pioneer attempt in machine translation in Brazil. (AU)