(This page is EUC-JP encoded)

Help for romaji to kana and kanji converter

Topics

• How to get kana?
• How to get small kana?
• Which are the available kana extensions?
• How to get punctuation marks and symbols?
• What is the semantic of quoted syllables?
• How to convert kana into kanji?
• How to conjugate verbal adjectives?
• How to conjugate ichidan verbs?
• How to conjugate godan verbs?
• How to get ascii and roman characters?
• How to get greek characters?
• Which characters are reserved in kana mode ?

How to get kana? [top] [topics]

Romaji is converted to kana according to the following tables. Capitalized syllables converts to katakana instead of hiragana. The case of non-first letters is ignored. Nihon, Halpern, and Kunrei romanization are all supported. Unmatched combinations leave a GETA-mark (thick equal sign) for each unconverted character.

How to get small kana? [top] [topics]

Small kana can be accessed explicitely with the usual x-notation. Some small kana have no equivalent in the hiragana table (the character index have no associated glyphs in the char-set), which is shown by a cross in the table, but will leave a blank in the output (because no glyph is associated to the corresponding char-index).

small kana set (JIS tables 0x24 and 0x25)
ぁ ァ	ぃ ィ	ぅ ゥ	ぇ ェ	ぉ ォ	ゃ ャ	ゅ ュ	ょ ョ	っ ッ	× ヵ	× ヶ	ゎ ヮ
xa	xi	xu	xe	xo	xya	xyu	xyo	xtu xtsu x	xka	xke	xwa

Generally, small kana are not called explicitely but are get via palatalative syllables and, as far as "xtsu" is concerned, via the germinative consonnants.

palatalative syllables
りゃ リャ	みゃ ミャ	ぴゃ ピャ	びゃ ビャ	ひゃ ヒャ	にゃ ニャ	ぢゃ ヂャ	ちゃ チャ	じゃ ジャ	しゃ シャ	きゃ キャ	ぎゃ ギャ
rya	mya	pya	bya	hya	nya	dya ja'	tya cha	zya ja	sya sha	gya	kya
りゅ リュ	みゅ ミュ	ぴゅ ピュ	びゅ ビュ	ひゅ ヒュ	にゅ ニュ	ぢゅ ヂュ	ちゅ チュ	じゅ ジュ	しゅ シュ	ぎゅ ギュ	きゅ キュ
ryu	myu	pyu	byu	hyu	nyu	dyu ju'	tyu chu	zyu ju	syu shu	gyu	kyu
りょ リョ	みょ ミョ	ぴょ ピョ	びょ ビョ	ひょ ヒョ	にょ ニョ	ぢょ ヂョ	ちょ チョ	じょ ジョ	しょ ショ	ぎょ ギョ	きょ キョ
ryo	myo	pyo	byo	hyo	nyo	dyo jo'	tyo cho	zyo jo	syo sho	gyo	kyo

A small-tsu for germinative consonnants is generated each time two consonnant of the following set are consecutive.

germinative consonnants
kk... gg... ss... zz... jj... tt... cc... dd... hh... ff... vv... pp... bb... ww... mm... rr...	っ... ッ...

Which are the available kana extensions? [top] [topics]

Some extra combinations are allowed for foreign syllables. They make sense only for katakana but are also accessable in hiragana.

extra syllables
		ふぁ ファ	×ぁ ヴァ
		fa	va
てぃ ティ	でぃ ディ	ふぃ フィ	×ぃ ヴぃ	うぃ ウぃ
ti'	di'	fi	vi	wi'
			× ヴ
			vu
		ふぇ フェ	×ぇ ヴェ	うぇ ウェ	ぢぇ ヂェ	ちぇ チェ	じぇ ジェ	しぇ シェ
		fe	ve	we'	je'	che	je	she
		ふぉ フォ	×ぉ ヴォ
		fo	vo

How to get punctuation marks and symbols? [top] [topics]

Japanese punctuation marks and symbols are all accessed by their names prefixed by a backslash. When an equivalent character exists in the ascii charset, the symbol can generally be accessed via this ascii-character (backslashed if it conflicts with another character).

punctuation marks and symbols (JIS table 0x21)
		、	。	，	．	・	：
		touten ,	kuten .	comma	period	ten	colon :
；	？	！	゛	゜	´	｀	¨
semicolon ;	question ?	exclam !	dakuten	handakuten	acute	grave	trema
＾	￣	＿	ヽ	ヾ	ゝ	ゞ	〃
circumflex ^	macron	underscore	kdou	vkdou	hdou	vhdou	nonoten
仝	々	〆	〇	ー	―	‐	／
asabove	dou	shime	maru	chouon -	hyphen2	hyphen	slash /
＼	〜	‖	｜	…	‥	‘	’
backslash	nyoro	bar2	bar |	tensen	2dots	lquote	rquote
“	”	（	）	〔	〕	［	］
lguillemet	rguillemet	lparen	rparen	lbracket2	rbracket2	lbracket [	rbracket ]
｛	｝	〈	〉	《	》	「	」
lbrace	rbrace	langle	rangle	langle2	rangle2	lkagi	rkagi
『	』	【	】	＋	−	±	×
lkagi2	rkagi2	lyomi	ryomi	plus +	minus	plusminus	times
÷	＝	≠	＜	＞	≦	≧	∞
div	eq =	neq	lt <	gt >	leq	geq	infty
∴	♂	♀	°	′	″	℃	￥
therefore	male	female	degree	prime	second	celcius	yen
＄	¢	£	％	＃	＆	＊	＠
dollar	cent	pound	percent %	hash	amper &	asterisk	at @
§	☆	★	○	●	◎	◇
section	star	blackstar	circle	blackcircle	circle2	diamond

extra symbols (JIS table 0x22)
	◆	□	■	△	▲	▽	▼
	blackdiamond	square	blacksquare	triangle	blacktriangle	downtriangle	blackdowntriangle
※	〒	→	←	↑	↓	〓
rose	yuubin	rightarrow	leftarrow	uparrow	downarrow	geta
		∈	∋	⊆	⊇	⊂	⊃
		in	ni	subseteq	supseteq	subset	supset
∪	∩
union	inter
		∧	∨	¬	⇒	⇔	∀
		land	lor	lnot	limp	leqv	forall
∃
exists
				∠	⊥	⌒	∂
				angle	bot	hat	partial
∇	≡	≒	≪	≫	√	∽	∝
nabla	equiv	dotfallingseq	ll	gg	sqrt	backsim	varpropto
∵	∫	∬
because	int	int2
			Å	‰	♯	♭	♪
			angstrom	perthousand	sharp	flat	note
†	‡	¶					◯
footnote	footnote2	paragraph					disc

What is the semantic of quoted syllables? [top] [topics]

Quoted syllables are syllables whose romaji representation clash with the romaji representation of another syllable. The unquoted syllable is the most frequently used. Use the quote to get the alternate syllable. Here are some interesting use of the quoted syllables:

• Provide the preferred romaji spelling for particles ("wa", "e", "o").
• The "n" syllable is quoted to block the greedy mecanism of the parser.
• "ji" (resp. "zu") syllable exists both as voiced "shi" (resp "su") and "chi" (resp. "tsu") syllables. We generally use the former except when the syllable follows a non-voiced "chi" (resp. "tsu"). In the latter case, use the quoted "ji" and "zu" syllables.

English	romaji input	kana output
The cat is drinking.	neko wa nondeimasu. neko wa' nondeimasu. neko ha nondeimasu.	ねこ わ のんでいます。 ねこ は のんでいます。
I go to the station.	eki e ikimasu. eki e' ikimasu. eki he ikimasu.	えき え いきます。 えき へ いきます。
I eat eggs.	tamago o tabemasu. tamago o' tabemasu. tamago wo tabemasu.	たまご お たべます。 たまご を たべます。
To be continued.	tsuzuku. tsuzu'ku. tsuduku.	つずく。 つづく。
To curl oneself up.	chiji ni maru. chiji' ni maru. chidi ni maru.	ちじ にまる。 ちぢにまる。
I like long distance swimming.	enei ga suki desu. en'ei ga suki desu.	えねいがすきです。 えんえいすきです。

How to expand kana into kanji? [top] [topics]

Kanji is supported in a rudimentary way by prefixing words by a star (*). A look-up is then performed in edict and the the first match is returned. A typical line in edict is an entry word written in kanji + kana, followed by a reading in kana between brackets.

What is searched in edict is the reading and what is returned is the entry. Usually, the reading is spelled in hiragana, but if some part of the reading is in katakana, you should also type this part in katakana in order to get the right match. E.g., for the entry "karaoke", the "kara" part is written in kanji and the "oke" part is written in katakana. Its reading use hiragana to spell "kara" and still use katakana for "oke", so you should type "*karaOKE" to get the match.

If the first match is not wanted, the n-th match (n from 1 to 99) can be accessed by the (n)-notation: just suffix the word by a one-digit number between parentheses. All alternatives are listed for the special value n=0. Alternatives are sorted as they are in edict, except that entries marked with the priority tag (P) are sorted separately and come grouped first.

romaji input	kanji output
*kanojo wa' *shinbun o' *yomu.	彼女は新聞を読む。
*watashi(0)	( 私(P) 渡し )
*watashi(1)	私
*watashi(2)	渡し
*watashi(3)	(none)
*watashi	私
*karaOKE	空オケ
*FURANSUgo	フランス語

Noun-adjectives (e.g shizukana) are listed in edict without their na-syllable. One must put a separator between the stem and the na-syllabe if we want to search a noun-adjective in edict. A space does the job, but is not visually a good solution. One can use the underscore (_) to suffix the stem of a noun-adjective (e.g. *shizuka_na) so that the stem is searched in edict. Underscores are separators just as spaces tabs and newlines.

How to conjugate verbal adjectives? [top] [topics]

Verbal adjectives (e.g. takai) are listed in edict with their i-form (stem + i-syllable). Use the tilde (~) after the stem of a verbal adjective to access its inflected forms (e.g. *taka~kunai). An i-syllable is concatened to the stem and a search in edict is performed. What is after the tilde is not used for the search. Note that the returned match is not necessarily a verbal adjective, and can be a homophone of another grammatical class. The (n)-notation takes place juste before the tilde.

romaji input	kanji output
*shinsetsu_na *hito desu yo.	親切な人ですよ。
kono *kaban wa' *aka~kunai.	この鞄は赤くない。

How to conjugate ichidan verbs? [top] [topics]

Ichidan verbs (e.g. taberu) are listed in edict with their infinitive form (stem + ru-syllable). Use the backquote (`) after the stem of an ichidan verb to access its inflected forms (e.g. *tabe`mashita). A ru-syllable is concatened to the stem, which restores the infinitive form (e.g. taberu) and a search in edict is performed. What is after the backquote is not used for the search. Note that the returned match is not necessarily an ichidan verb, and can be a homophone of another grammatical class. The (n)-notation takes place juste before the backquote.

How to conjugate godan verbs? [top] [topics]

Godan verbs (e.g. nomu) are listed in edict with their infinitive form (stem + utsuru/munubu/ku/gu/su-syllable). Use the double-quote (") after the stem of a godan verb to access its inflected forms. An infinitive ending-syllable is concatened to the stem and a search in edict is performed. The ending-syllable is determined by one or two syllables following the double-quotes (e.g *no"ma/mi/me/mu all restore the infinitive form nomu). The (n)-notation takes place juste before the double-quote. For past an progressive forms, the two syllables just after the colon are used to restore the infinitive form for the edict-search. The engine looks at "nda/"nde to restore munubu-verbs, at "tta/"tte for utsuru-verbs, at "ita/"ite for ku-verbs, at "ida/"ide for gu-verbs, and at "shita/"shite for su-verbs. Note that the returned match is not necessarily a godan verb, and can be a homophone of another grammatical class.

romaji input	kanji output
*niku o' *tabe`ta.	肉を食べた。
BI-RU o' *no(2)"manai.	ビールを飲まない。
BI-RU o' *no(2)"nda.	ビールを飲んだ。
*kutsushita o' *ka"tta.	靴下を買った。
*tegami o' *ka"ita.	手紙を書いた。
*umi de *oyo"ida.	海で泳いだ。
*nihongo de *hana"shita.	日本語で話した。

Exceptionnal verbs: not supported yet.

How to get ascii and roman characters? [top] [topics]

The ascii mode allows to block the kana and kanji conversion. Ascii mode is entered and leaved by a dollar ($). Ascii mode can be used to type plain ascii or HTML tags, Yet, no dollar can be typed in ascii mode.

Double-sized roman alphabet and numerals are accessed in kana-mode via the sharp (#). Each roman word or number must be prefixed by the sharp.

Roman letters and numerals (JIS table 0x23)
０	１	２	３	４	５	６	７	８	９
0	1	2	3	4	5	6	7	8	9
ａＡ	ｂＢ	ｃＣ	ｄＤ	ｅＥ	ｆＦ	ｇＧ	ｈＨ	ｉＩ	ｊＪ
a	b	c	d	e	f	g	h	i	j
ｋＫ	ｌＬ	ｍＭ	ｎＮ	ｏＯ	ｐＰ	ｑＱ	ｒＲ	ｓＳ	ｔＴ
k	l	m	n	o	p	q	r	s	t
ｕＵ	ｖＶ	ｗＷ	ｘＸ	ｙＹ	ｚＺ
u	v	w	x	y	z

How to get greek characters? [top] [topics]

Greek alphabet is accessed by prefixing the name of each letter by a backslash (\). If the first letter of the name is uppercase, then the resulting greek letter is uppercase instead of lowercase.

Greek alphabet (JIS table 0x26)
α Α	β Β	γ Γ	δ Δ	ε Ε	ζ Ζ	η Η	θ Θ
alpha	beta	gamma	delta	epsilon	zeta	eta	theta
ι Ι	κ Κ	λ Λ	μ Μ	ν Ν	ξ Ξ	ο Ο	π Π
iota	kappa	lambda	mu	nu	xi	omicron	pi
ρ Ρ	σ Σ	τ Τ	υ Υ	φ Φ	χ Χ	ψ Ψ	ω Ω
rho	sigma	tau	upsilon	phi	chi	psi	omega

Which characters are reserved in kana mode? [top] [topics]

The following characters are reserved in kana-mode. Any ascii-character is allowed in ascii-mode except the dollar ($).

reserved character	function
' quote	alternate kana conversion for some syllables.
* star	expansion of kana into kanji.
(n) parentheses and digits	alternate kanji expansion.
_ underscore	separator used, e.g., for na-adjectives.
~ tilde	verbal adjective conjugation.
` backquote	ichidan verb conjugation.
" double-quote	godan verb conjugation.
# sharp	prefix for roman words and numerals.
\ backslash	prefix for greek letters and symbols.
$ dollar	enter and leave ascii mode.