Welcome to the Sámi language technology project

A Northern Sámi wordform like e.g. sáni may be genitive or accusative singular of the word sátni, which means "word". In the former case, it may be a noun modifier (sáni mearkkašupmi, "the word's meaning") or a postposition complement (sáni birra "about the word"), in the latter, it is an object (dovdan dan sáni "I know that word"). By grammatical analysis we mean a process that assigns all possible morphological an syntactic properties to a given wordform, and by disambiguation, we mean a process that determines which of the available options is the appropriate one in a given context.

Wordform analysis is useful when one wants to find the baseform of any given wordform. The form humat may e.g. be an inflectional form of any of the words hupmat, humahit, humadit, hupma. Disambiguation is useful for text analysis. In order to find out the use of the word sátni as an object, one first has to be able to distinguish the accusative occurrences from the genitive ones.

Word generation means the generation of wordforms, on the basis of a lexeme and a grammatical specification. The wordform generator can tell that the indicative present first person singular of boahtit (to come) is boađán.

Wordform analysis is useful in pedagogical applications, where a student e.g. may wonder what is the present tense first person dual of the same verb (it is bohte). It is also a basic component in machine translation systems, in order to translate "I come" to Northern Sámi, one needs a process for going from boahtit to boađán.

During the project we have made small webapps where you are able to test our technology interactively.

In the menu to the left, you can disambiguate and analyse Northern Sámi words and sentences, and analyse Lule and South Sámi words. You will also have the opportunity to generate Northern, Lule and South Sámi words and numerals.

Last revision: $Date: 2011-06-29 20:05:09 +0200 (ons, 29 jun 2011) $, by $Author: trond $