Mansi documentation
{
"type": "Feature",
"properties": {
"name": "Mansi",
"radius": 200,
"marker-color": "#ff4444",
"marker-size": "large"
},
"geometry": {
"type": "Point",
"coordinates": [61.344, 64.4118]
}
}
Center location data taken from Glottolog. Area extent is local data. Both can be adjusted if wrong - file a pull request!
These pages document the work on the Mansi language model.
The Mansi analyser contains appr 6000 stems, and is in an early stage. It still recognise 48 % of the wordforms in a 1.1 million corpus of running text and is in use in a Mansi e-dictionary. For questions, contact Csilla Horvath, Jack Rueter or Trond Trosterud.
Documentation and resources
HOWTO for developers
- How to download, set up, compile and use the language model
- A list of the morphological tags used in the GiellaLT infrastructure
Resources built upon the Mansi language model
- Language technology tools for Mansi (based upon this model)
- xperimental Mansi spellchecker for download
- Experimental Mansi spellchecker online (click ABC and choose Languages)
The mns project
Documents
Testing diaries
Meeting memos
-
2024: 30.1. ◊ 8.2. ◊ 16.2. ◊ 23.2. ◊ 20.3. ◊ 7.4. ◊ 11.4. ◊ 22.4. ◊ 29.4. ◊ 6.5. ◊ 5.6. ◊ 27.6. ◊ 14.8. ◊ 19.9. (with Kildin) ◊ 13.11.
-
2023: 2.6. ◊ 9.6. ◊ 16.6. ◊ 27.6. ◊ 1.9. ◊ 12.10. ◊ 17.11. ◊ 29.11. ◊ 12.12.
In-source documentation
Below is an autogenerated list of documentation pages built from structured comments in the source code. All pages are also concatenated and can be read as one long text here.
src/cg3/fst/- mansi-lexicon.lexc (src)
morphology/affixes/- phonology.twolc (src)
- root.lexc (src)
stems/
phonetics/transcriptions/
tools/grammarcheckers/tokenisers/