hiroshi-manabe / ngram-converter-cpp

Kana-Kanji (and vice versa) converter using N-gram data (C++ version, dictionary data included)

ngram-converter-cpp

A simple N-gram Kana-Kanji/Kanji-Kana converter (C++ version).

The attached dictionary files are converted from the files that can be downloaded at http://plata.ar.media.kyoto-u.ac.jp/gologo/lm.html .

You will need "marisa-trie" module to use this software.


You can test the conversion program in the following way:

1. Make the target.
> make

2. Run the program with the attached dictionary.

From Kana to Kanji:
> ./converter-main -i dict/bccwj4_rev

From Kanji to Kana:
> ./converter-main -i dict/bccwj4


To build a dictionary from SRILM format N-gram file, run sort-ngram.py and generate-data.py in this order.