Difference between revisions of "Ideas for Google Summer of Code/Accent and diacritic restoration"

From Apertium
Jump to navigation Jump to search
Line 7: Line 7:
 
* Allow rule-based replacements of character sequences.
 
* Allow rule-based replacements of character sequences.
 
* Train models for all languages in Apertium.
 
* Train models for all languages in Apertium.
  +
* Inform charlifter with target-language information from a target-language model.
* ...
 
   
 
==Coding challenge==
 
==Coding challenge==

Revision as of 14:05, 14 March 2013

Many languages use diacritics and accents in normal writing, and Apertium is designed to use these, however in some places, especially for example. instant messaging, irc, searching in the web etc. these are often not used or untyped. This causes problems as for the engine, traduccion is not the same as traducción. Create an optional module to restore diacritics and accents on input text, and integrate it into the Apertium pipeline.

Tasks

  • Finish the port of Kevin Scannell's charlifter to C++, it should respect superblanks
  • Allow rule-based replacements of character sequences.
  • Train models for all languages in Apertium.
  • Inform charlifter with target-language information from a target-language model.

Coding challenge

Frequently asked questions

See also