Samest meeting 9.6.2017

Present: Fran, Heiki, Heli, Jaak, Jack, Sjur, Sulev, Trond


  • Summary from Gothenburg Samest meeting
  • Happened since last time
  • Future

Summary from Gothenburg Samest meeting

  • future projects - võro fst should be finalised (Sulev, Jack, Heiki, Heli?)
  • intent to apply for funding from EKT (Estonian Language Technology programme)
  • cooperation project on iCALL with Uppsala University? (where Estonian is taught) - Heli has written an e-mail to prof Raimo Raag but not got any answer yet

Happened since last time

  • sme2est (Käbi Suvi's BA thesis); she got "B" https://svn.code.sf.net/p/apertium/svn/incubator/apertium-sme-est
  • Võro keyboards are created by Sjur for Windows and Mac Tested both on Mac and Windows - works good. Windows installer could be better.
  • Heiki has worked with jompikumpi -type words in est (double inflecting words), e.g. emb-kumb, kihin-kahin, seesama, kakssada. This done with concatenating two transducers.
  • Fin-Est dict (Kotus) from EKI to be used in Apertium

Testing with sms and izh, but things should be ready for expanding this summer to other languages.

Testing xml output for finsms:

http://sanat2.csc.fi:8000/smsxml/xml_out/?language=sms&type=finsms&file=V_finsms.xml xml output for _sms2x.xml

http://sanat2.csc.fi:8000/smsxml/xml_out/?language=sms&type=morph&file=V_sms2x.xml xml output for ped/sms/src/*xml


FST speller for predictive input

Documentation: https://giellalt.uit.no/keyboards/PlanForMoreKeyboards.html

Possible extensions:

  • grammar checking
  • predictive word completion (based upon start of word)
  • predictive next word (based upon last word(s))

Next Võro meeting (Heli, Jack, Sulev) will be 20.06. Then we can discuss closer what you have talked in Göteborg (future of Võro FST, funding) and other things.


  • Immediate
    • Estonian fst in shape
    • fin-est dictionary into bidix format
  • X-fin
    • Improve fin doubleforms and strange forms
  • Project closing conference 21-22.9. in Laulasmaa, we should write a comprehensive presentation of what we did.

Future funding:

Next meeting

Tuesday 29. August 13:00 Norwegian time