Cree Corpus: A Collection of nêhiyawêwin Resources

Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics Volume 1: Long Papers, pages 6354 - 6364 May 22-27, 2022 c⃝2022 Association for Computational Linguistics Cree Corpus: A Collection of nêhiyawêwin Resources Daniela Teodorescu, Josie Matalski, Delaney Lothian, Denilson Barbosa, Carrie Demmans Epp Dept. of Computing Science University of Alberta {dteodore,jmatalsk,dlothian,denilson,cdemmansepp}@ualberta.ca Abstract Plains Cree (nêhiyawêwin) is an Indigenous language that is spoken in Canada and the USA. It is the most widely spoken dialect of Cree and a morphologically complex language that is polysynthetic, highly inflective, and agglutinative. It is an extremely low resource language, with no existing corpus that is both available and prepared for supporting the development of language technologies. To support nêhiyawêwin revitalization and preservation, we developed a corpus covering diverse genres, time periods, and texts for a variety of intended audiences. The data has been verified and cleaned; it is ready for use in developing language technologies for nêhiyawêwin. The corpus includes the corresponding English phrases or audio files where available. We demonstrate the utility of the corpus through its community use and its use to build language technologies that can provide the types of support that community members have expressed are desirable. The corpus is available for public use1. 1 Introduction Recent work with Indigenous persons has shown that some want advanced technologies to support the learning and use of their languages. The Cree and Métis persons involved in this study stated a desire for technologies such as an app to help with learning the structure of the language for conversation, translation, and AI agents that re- semble a speaker (Lothian et al., 2019). Partic- ipants wanted these tools to support interaction in nêhiyawêwin (Plains Cree) or the learning of this language. All of these larger ideas are depen- dent on core language technologies such as lan- guage models, speech recognition, speech synthe- sis, or machine translation. However, a lack of 1https://github.com/EdTeKLA/ IndigenousLanguages_Corpora publicly available corpora hinders the development of such technologies for low-resource languages like nêhiyawêwin. Government policies have contributed towards supporting the preservation and revitalization of some Indigenous languages, e.g., Inuktitut (Joa- nis et al., 2020). However, many have not bene- fited from this level of support for developing re- sources and technologies. Recently, some govern- ment informational material such as voter guides or COVID-19 pamphlets have been translated into nêhiyawêwin. Nevertheless, the availability of re- sources is still limited and short texts or other re- sources are distributed across libraries and the Inter- net. To understand why this is the case, we need to reflect on the colonial practices that have attempted to eradicate a language and people. Previous and on-going government policies and practices, such as the implementation of residential schools (Bom- bay et al., 2011), have left a small number of fluent speakers and language resources for nêhiyawêwin- speaking communities. These practices prevented and continue to pre- vent the development of language technologies be- cause state-of-the-art statistical and neural models require large amounts of text. To work …

Content truncated to 500 words