JOS ToTaLe text analyser for Slovene texts

Note that this is a very old service, and better web tools for tagging Slovene exists, in particular ReLDIanno.

Here you can Tokenise, Tag and Lemmatise Slovene texts. The tags (morphosyntactic descriptions, MSDs) follow the JOS morphosyntactic specifications and can be shown either in Slovene (e.g. Gp-g = glagol pomožni pogojnik) or English (e.g. Va-c = Verb auxiliary conditional). The output file is in "vertical" format, appropriate for using in SketchEngine and CWB. Each line is either an XML tag (<doc>, <p>, <s> and </s>, </p>, </doc>) or an annotated token. Token lines are tab-separated and contain 1) the token, 2) the lemma (base form) of the word, and 3) the MSD tag. For punctuation, the MSD and lemma fields are identical to the token. The MSDs can be converted into various other formats with the JOS MSD conversion tables.

Analyse the text and the result or the compressed files, with the MSDs in Slovene or English.

Type or paste text into the window below:
Pričujoči spletni servis vam vrne vaše besedilo, razdeljeno na stavke in pojavnice, pri čemer je vsaka beseda označena s svojo lemo in oblikoskladenjsko oznako JOS. Besedilo lahko prilepite v pričujoče okno ali pa ga naložite kot datoteko z vašega računalnika. Datoteka mora biti golo besedilo (.txt), zapis znakov pa utf-8 ali pa koprimirana datoteka (.zip, .rar, .tgz) z datotekami .txt. Velikost datoteke ni posebej omejena, vednar pa je praktična zgornja meja okoli 500.000 besed. Rezultat pretvorbe lahko pogledate ali pa prenesete kot stisnjeno datoteko (.zip) na svoj računalnik. POZOR: Pri označevanju prihaja tudi do napak. Nekaj jih lahko najdete že, če označite tole besedilo...
or upload plain utf-8 text files: (.txt, or .zip, .rar, .tgz)

the form!
Note: uploaded files are being archived, and could be used as a basis for further research.

Documentation

ws-jos-ljubljana-06.txt: Grammatical relations file (using Slovene MSDs), needed for compiling Word Sketches in SketchEngine.
The grammatical relations for JOS MSDs are based on the relations developed for the FidaPLUS corpus by Simon Krek and described in Simon Krek, Adam Kilgarriff: Slovene Word Sketches. Proceedings of 5th Slovenian and 1st international Language Technologies Conference 2006 Jozef Stefan Institute, Ljubljana, Slovenia.
Old paper describing ToTaLe:
Tomaž Erjavec, Camelia Ignat, Bruno Pouliquen, Ralf Steinberger. Massive multi-lingual corpus compilation: Acquis Communautaire and totale. In Proceedings of the 2nd Language & Technology Conference, April 21-23, 2005, Poznan, Poland. 2005, pp. 32-36.
Papers describing the training set for ToTaLe used in this service:
Tomaž Erjavec, Simon Krek: The JOS Morphosyntactically Tagged Corpus of Slovene. Proceedings of the Sixth International Conference on Language Resources and Evaluation (LREC'08), Marrakech 2008.
Tomaž Erjavec, Darja Fišer, Simon Krek, Nina Ledinek. The JOS Linguistically Tagged Corpus of Slovene. Proceedings of the Seventh International Conference on Language Resources and Evaluation (LREC'10), Malta, 2010.

Related services

ReLDIanno
Obeliks
LemmaGen: multilingual lemmatisation
NoSketch Engine: tagged corpora avialiable through the open source version of SketchEngine

Page last updated 2022-05-14, et