Clone wiki

libkkc / Home

What's this?

libkkc provides a converter from Japanese Kana-string to Kana-Kanji-mixed-string. It was named after kkc.el in GNU Emacs, a simple Kana Kanji converter, while libkkc tries to convert sentences in a bit more complex way using N-gram language models.


Note: since 0.1.3, marisa-trie Python binding (included in marisa-trie) is needed to install libkkc-data.

  1. compile and install marisa-trie
  2. checkout or download libkkc and ibus-kkc and install them
  3. download and install libkkc-data


First, please check if the basic functionalities are working with kkc command. See the example in README.

If it seems to work, then you can proceed to set up the IBus engine. If you are using GNOME 3.6, you will need to disable the IME whitelist with:

$ gsettings set org.gnome.desktop.input-sources show-all-sources true

though this is not needed under GNOME 3.8. Then you will be able to add "Japanese (Kana Kanji)" input source from gnome-control-center "Region and Language" panel.

For other desktops, use ibus-setup tool to add it.

Key bindings

  • space: convert sentence or segment
  • Control-g or Control-c: cancel conversion
  • Right: move to the next segment
  • Left: move to the previous segment
  • Shift-Right: expand the current segment
  • Shift-Left: shrink the current segment
  • Tab: complete word
  • select region, Alt-r, and then convert sentence: add word to the dictionary
  • Control-BackSpace: remove word from the dictionary
  • Alt-k: cycle kana input modes
  • Alt-l: cycle latin input modes
  • Alt-grave or Alt-@: toggle direct input mode
  • F6: convert the current preedit to Hiragana
  • F7: convert the current preedit to Katakana
  • F8: convert the current preedit to Halfwidth Katakana
  • F9: convert the current preedit to Fullwidth Latin
  • F10: convert the current preedit to Latin
  • Muhenkan: cancel the current conversion

Bug reporting and discussion

File any issues here. There is also a discussion mailing list (in Japanese or English).

Have fun!