Introduction

L3XDG will, in the near future, use some machine learning techniques to try to offset the ambiguity that comes with using a rule-based system, exploring the most probable translations first.

In order to facilitate this, we'll make use of the small amount of bitext that's available for our languages of interest.

Quechua corpora

bitext

• AvenueQuechuaCorpus
• Quechua/Spanish text shared with us by the SQUOIA Project Team! Quechua Spanish
• The Peruvian Constitution. (Spanish with interleaved Quechua translations.)
• Catholic bibles, pointed out to us by Rada Mihalcea and her team.
• Quechua - Cuzco
• Quechua - Bolivia
• Spanish
• Mike converted the dictionary from runasimi.de to XML (it was a big spreadsheet before). http://www.cs.indiana.edu/~gasser/L3/runasimi.xml

dictionaries

Quechua dictionaries are available...

• http://www.runasimipi.org/blanco-en.php?file=diccionarios

relevant information

Dialect codes for the regional dialects of Quechua are listed here: http://en.wikipedia.org/wiki/ISO_639:q

Guaraní / Jopará corpora

• http://www.staff.uni-mainz.de/lustig/guarani/art/jopara.pdf
• http://www.cabildoccr.gov.py/libros/Ramona%20Quebranto%20(LIBRO).pdf