Asheninka (a Syllable Parser) is a tool to explore using several different algorithms for parsing orthographic words into syllables. It can also insert discretionary hyphens in words. See It is designed to work with any language whose orthography is roman-based.

It has the following capabilities:

  • One can get word lists from ParaTExt and FLEx.
  • One can see predicted syllabification and quickly compare to correct syllabification.
  • One can export hyphenated words for use by ParaTExt and InDesign.
  • Runs on Linux, Mac,and Windows.
  • It can syllabify thousands of words per second. (Over 50,000 words/second on my computer.)
  • One can backup and restore data sets. The backup can include a comment about what the state was at the time.
  • Can compare what is different between two backups (or the current implementation and a backup)
  • It can be used as a spelling checker (if a word will not syllabify, it may have a typo). (Noted by David J. Weber in

Version has the following improvements:

  • There is User Documentation available from the Help menu item / User Documentation.
  • The parser result field now attempts to give an indication of how far the process got when it failed.
  • A given segment can now be used in more than one natural class.
  • It now remembers what view and which item in that view you were last in upon exit. So the next time it will come back to that same place.
  • It now remembers the size of the main window.
  • Various bug fixes.

Version Alpha

This is an alpha version. We expect it to have bugs and it is incomplete. Nonetheless, it can still be used for useful, productive work. Be sure to use the “File menu item / Project Management / Back up this project” capability. Note that you can insert a comment when backing up and thereby remember what was special about that backup.

Known Issues

  • Only the CV Pattern approach is implemented
  • When first starting a view/tool and there are no items, still shows the fields in the right pane.
  • When delete an item from CV Patterns, still shows one that should not be there.
  • The Spanish user interface probably has many errors in it. Any corrections are welcome.