Rust Kana Transliterator

This project transliterates English and Japanese words into Japanese characters, and can also take Japanese characters and output the corresponding romanization.

Open source license.

Support for transliterating into and from both hiragana and katakana is included.

Usage

To run the program, you must input cargo run [OPTION] [INPUT] into the command line. For example, to generate the hiragana for "gakkou", you would input:

cargo run hiragana gakkou

This will output the input text in hiragana:

    Finished dev [unoptimized + debuginfo] target(s) in 0.07s
     Running `target/debug/to_kana hiragana gakkou`
がっこう

When an input contains whitespace, make sure to surround it with quotation marks:

cargo run katakana "       ka ta ka naaaaaaaaa "
       カ タ カ ナーーーーーーーー

Options

This program has six options.

hiragana

This option expects a Japanese input string and will output a hiragana string. While the input does not necessarily need to be a Japanese word, it should still follow Japanese phonological rules.

Example:

cargo run hiragana "jibirisu"
じびりす

When inputting a Japanese word, this program will accept modern Hepburn romanization practices with some exceptions:

• Since the program can only accept ASCII input, long vowels will be doubled rather than using a macron. For instance obāsan should be input as obaasan.
• Geminates are input by repeating the doubled character in all cases. This differs from Hepburn romanization where まっちゃ is usually romanized as matcha. For this program use maccha.
• To resolve the abiguity of final -n (ん) and the initial ny- sound (as in nya にゃ), ny- digraphs should be input as nq-. All other digraphs will follow standards. For example:

cargo run hiragana "honya myuu jaa"
ほんや みゅう じゃあ
cargo run hiragana "honqa myuu jaa"
ほにゃ みゅう じゃあ

katakana

This option accepts a Japanese input string in Latin 1 characters and will output a katakana string. The input requirements are the same for the hiragana option.

cmu_hiragana

This option will take an English word as an input and transliterate to a hiragana string output. There are two methods used: a rules-based method which parses english words, and a syllable-generator which uses the Carnegie Mellon University phonetic dictionary (http://www.speech.cs.cmu.edu/cgi-bin/cmudict) to first create phonetic syllables, which can then be transliterated into Japanese, hopefully giving a result closer to the actual pronunciation.

Example:

cargo run cmu_hiragana "rust"
らすと

Note that this option can only take one word at a time as an input.

cmu_katakana

This option functions the same as cmu_hiragana but will output a katakana string.

Example:

cargo run cmu_katakana "rust"
ラスト

roomaji_hiragana

This option will take a Japanese hiragana string and will output it as Latin 1 unicode.

Example:

cargo run roomaji_hiragana らすと
rasuto

romaji_katakana

This option will take a Japanese katakana string and will out it as Latin 1 unicode.

Example:

cargo run roomaji_katakana ラスト
rasuto