Avenue Project

The project's team
Current Projects:   Aymara (Bolivia);   Quechua (Peru);   Mapudungun (Chile).


Machine Translation (automatic translation of one language into another by computer) is rapidly becoming available. Many free translation servers are available on the Web and many commercial products are also available for purchase. The availability of Machine Translation (MT) promises to make more information available to more people, while at the same time, allowing local languages to flourish by reducing the need to speak a common language. However, a little known fact about MT is that many of the current programs require on the order of a person-century of human effort to develop or, alternatively, require well over 200~MB of high-quality text in electronic form plus up to a person-decade of development time. Such high requirements in terms of human effort and electronic data render current MT infeasible for all but a handful of "major" languages (such as English, Spanish, and Japanese). Economic considerations may have the dual effect of disenfranchising large linguistic minorities and economically developing nations from full participation in the information age, while preventing governments, companies, policymakers, and researchers access to ideas and opportunities in much of the developing world.

The Avenue project has both social and scientific goals in Machine Translation. The scientific goal of Avenue is to explore methods to reduce the development time and cost of MT, thereby making it available to more people. The social goal is to contribute to a growing trend to reinstate indigenous languages in official uses outside of the home, and to prevent the disenfranchisement of speakers of indigenous languages. Avenue may contribute to this goal by providing access to government documents in indigenous languages, by including monolingual speakers of indigenous languages in surveys of health and agriculture, or by contributing to bilingual multicultural education. An important aspect of our approach is to rely on indigenous communities, competent native speakers and local scientists to design and develop projects that are relevant to their own plans for local and community development.

History  /  Cyber Tools for Amerindian Languages /  Relevant Papers & Publications