Background information on the South Saami project

Work on South Saami today

Giellatekno and Divvun work on South Saami within several different projects:

• The basic work: The South Saami morphological and syntactic analysers
• The South Saami spellchecker
• The Åarjelsaemien-daaroen digibaakoeh
• The South Saami OAHPA pedagogical program

The initial South Saami project (1998-1999)

South Saami was the first Saami language to receive an analysis within the Saami language technology project. Sjur Moshagen and Trond Trosterud made a morphological parser for South Saami already in 1998, before the Giellatekno was initiated. The parser included a morphological analysis of the case and number inflection of nouns, without derivation and compounding.

The original Moshagen and Trosterud files are found at Lingsoft (in the sms directory there). They wrote an article on that parser, cf. the link below. Later, they received some comments from Lauri Karttunen that they (at some point) will incorporate in the article. The present sms parser is based upon Karttunen’s critique. Contrary to the original version, it is able to generate forms as well as analysing them. As part of the evaluation of Karttunen’s comments, the southern Saami lexicon has been transformed from Lingsoft format to Xerox format.

Later, the parser was enlarged to include handling of the closed parts of speech. We also did some attempts to include verbal inflection, but we did not succeed in analysing the South Saami verbs within the same framework as for the nouns.

The article A Two-level Parser for South Saami, is the most complete documentation of the first phase of the South Saami parser.

The South Saami spell checker project (2008-2010)

The main focus of the Divvun 2 project was the construction and release of a South Saami spellchecker. The spellchecker was released in december 2010.

Phase 2: The South Saami spell checker project

The main focus of the Divvun 2 project was the construction and release of a South Saami spellchecker. The spellchecker was released in december 2010.

Update from Nov 4, 2021

Sørsamiske feil som dominerer:

• suffiksfeil
• vokalfeil (omlyd)
• preaspirasjon/geminerte plosivar
• fotstruktur (vokalar som for likestava når det er trestava)
• to eller fleire feil
• låneordstilpassing (særleg vokallengde)