PYKAKASI documentation

This is the documentation for Pykakasi library and utility. pykakasi is a library and utility implemented KAKASI functionality in Python. KAKASI was originaly built to convert Japanese text to roman form.

pykakasi is a free software, and available on GitHub project.

wakati is an implementation of kakasi's wakati gaki option.

Command Line Options

Command line options

Programming Interface

Application programming interface and options

Pykakasi authors

PyKAKASI authors and credits

Copyright and License

Copyright and license

Table of contents

Access to all of document contents

Supported python versions

Pykakasi supports python 2.7, python 3.5, 3.6, 3.7, 3.8 and PyPy.

It may work with python 2.6, 3.3, 3.4 and pypy3 but these are not tested now.


It depends on klepto for providing a mapping database.


KAKASI is the language processing filter to convert Kanji characters to Hiragana, Katakana or Romaji 2 and may be helpful to read Japanese documents.

The name "KAKASI" is the abbreviation of "kanji kana simple inverter" and the inverse of SKK "simple kana kanji converter" which is developed by Masahiko Sato at Tohoku University. The most entries of the kakasi dictionary is derived form the SKK dictionaries. If you have some interests in the naming of "KAKASI", please consult to Japanese-English dictionary. :-)

Table of contents

Indices and tables



"Romaji" is alphabetical description of Japanese pronunciation.