romaji

romaji is a converter library to romanize Japanese hiragana/katakana string by standard and IME typing style. Even though pykakasi already has provided the same functions, and returns only one romanized string. However, there exists several different romanization systems, so one hiragana/katakana string has so many romanize string. For example, "ちゃ" can be romanized as "cha", "tya", "chixya", "tixya", "chilya", or "tilya". romaji provides romanized strings as many as possible.

If an input string contained non hiragana/katakana characters (includes kanji), romaji return the characters as same as the input. For example, romaji converts the input "お茶の水" to "o茶no水". If there is a need to romanize the whole string which includes kanji, romaji can romaize the readings in the tokens which Janome tokenizer provides. romaji strongly recommends to use with Janome.

The mapping from hiragana/katakana to romaji is based on common IME's system to input Japanese. Therefor, romaji does not directly implement the standard system like Hepburn, Nihon-shiki or Kunrei-shiki, but includes them.

Installation

$ pip install romaji

Usage

>>> import romaji
>>> romaji.transliterate('僕ドラえもん')
['僕doraemon', '僕doraemon\'', '僕doraemonn']
>>> romaji.transliterate('諸行無常')
[]