You can use these suggested parameters to increase the accuracy of tesseract.
example usage from my jtesseract script:
tesseract "$@" -l jpn -c chop_enable=T -c use_new_state_cost=F -c segment_segcost_rating=F -c enable_new_segsearch=0 -c language_model_ngram_on=0 -c textord_force_make_prop_words=F -c edges_max_children_per_outline=40
Jakaroma does this, and it is based on Kuromoji which is much better at kanji transliteration than Kakasi and even Mecab.
Open source: Apache 2 license
Size: 11 MB
Transliteration quality: High
Disclaimer: Maintained by me.
Kakasi Java does exactly this.
It also includes a command-line tool so that you can easily try it.
Open source: GNU-GPLv2 license
Size: 3.2 MB
Transliteration quality: Poor
Disclaimer: Maintained by me.
Note: Mecab is more accurate at guessing pronunciations but is not 100% Java unfortunately.
I've written a few sed-based filters for this purpose.
Example converting Romaji to Kana:
romaji2kana <<< arigatou
More binary alternatives i know:
uconv -x 'Any-Hiragana' or uconv -x 'Any-Katakana'
kakasi (untested by me)
mecab (untested by me)
According to http://ipad-ssm.net/archives/211 there's an option to turn on the 10 key instead of having to use your 55 key. It's tricky to find this on your own though.
To do this, you need to have the "Split Keyboard" option on, found in Settings -> General -> Keyboard
Proceed to click on "International Keyboards" and add a new keyboard. You can select the ...
I think I found it out myself: it is NTT DOCOMO JSpeak.
Available for iOS and Android. Free for five uses then you have to pay for a time-based subscription (1 week, 2 weeks etc.)
Important notice: apparently it requires a SIM, so you cannot use it on a Tablet or from a SIM-less phone (the SIM does not need to be Japanese, though).
The one that was introduced by my 先輩 since the first time I learnt Japanese 6 years ago is WWWJDIC. It's so famous that a lot of online dictionaries I found use the database from WWWJDIC
By default it'll search in the general dictionary (EDICT) but you can switch to the Japanese name dictionary (ENAMDICT) for more detailed information about names
I have found http://kanji.reader.bz to be rather adapted to this use. Enter the kanji given name in the input field, press the button and you get a list of possible pronunciations:
Seems to be based on current names
Limited to people names, so results are not polluted by common words.
Not indication of how popular a name is.
I don't know any good open source software for this, but if that is not a hard requirement, I have used ABBYY FineReader 9.0 Express Edition for a couple of years now, and I am generally pleased with the quality of the OCR results. Not sure if I can say better than 98% but good enough to be easily correctable.
I use it mostly to OCR Japanese text from books ...
You can actually use function on google translate to "draw" the symbols. It's available on the app, and I believe it might be available on the website. I found this incredibly useful when first learning the Korean alphabet, as it will recognize symbols or their closest match. Good luck!
Mecab is pretty accurate for this.
It is open source.
On Ubuntu 2015.10:
sudo apt-get install mecab libmecab-dev mecab-ipadic mecab-ipadic-utf8
Yes you ned all of these packages, the mecab package alone won't do.
echo 東京特許許可局 | mecab