The 46 basic hiragana.
Follow a small group across the chart, then hide the romaji and read it in a different order.
| a | i | u | e | o |
|---|---|---|---|---|
| あa | いi | うu | えe | おo |
| かka | きki | くku | けke | こko |
| さsa | しshi | すsu | せse | そso |
| たta | ちchi | つtsu | てte | とto |
| なna | にni | ぬnu | ねne | のno |
| はha | ひhi | ふfu | へhe | ほho |
| まma | みmi | むmu | めme | もmo |
| やya | ゆyu | よyo | ||
| らra | りri | るru | れre | ろro |
| わwa | をo | |||
| んn |
This is the basic set. Marks such as dakuten, small kana and long sounds come next. を is shown as o for its usual pronunciation; type wo to enter it. は and へ are pronounced wa and e when used as particles.
Compare with the Japan Foundation’s kana reference (PDF) ↗Small marks. More sounds.
Keep these references beside the basic chart. The readings use common romaji spellings; learn each form inside a word too.
Voiced and p-sounds25 forms · dakuten & handakuten
| a | i | u | e | o |
|---|---|---|---|---|
| がga | ぎgi | ぐgu | げge | ごgo |
| ざza | じji | ずzu | ぜze | ぞzo |
| だda | ぢji | づzu | でde | どdo |
| ばba | びbi | ぶbu | べbe | ぼbo |
| ぱpa | ぴpi | ぷpu | ぺpe | ぽpo |
じ / ぢ and ず / づ share the common readings shown here but remain different spellings. Learn the spelling with the word.
Contracted sounds33 combinations · small ゃ・ゅ・ょ
| ya | yu | yo |
|---|---|---|
| きゃkya | きゅkyu | きょkyo |
| しゃsha | しゅshu | しょsho |
| ちゃcha | ちゅchu | ちょcho |
| にゃnya | にゅnyu | にょnyo |
| ひゃhya | ひゅhyu | ひょhyo |
| みゃmya | みゅmyu | みょmyo |
| りゃrya | りゅryu | りょryo |
| ぎゃgya | ぎゅgyu | ぎょgyo |
| じゃja | じゅju | じょjo |
| びゃbya | びゅbyu | びょbyo |
| ぴゃpya | ぴゅpyu | ぴょpyo |
The small character joins the preceding one. Compare きゃ (kya) with きや (ki-ya). These are common combinations, not every possible spelling.
Look closely. Read the whole form.
Move from single characters to short words. Choose one reading; an explanation follows every answer. At the end, revisit the ones you missed.
Which reading matches?
し
All ten readings and explanations
し
shi — し is shi. Keep it distinct from ち (chi) and つ (tsu).
ぬ
nu — ぬ is nu. Compare its ending with め (me), then try recognising it on its own.
れ
re — れ is re. Read it outside its chart row so the position does not give you the answer.
が
ga — The dakuten changes か (ka) to が (ga). The small marks belong to the character.
ぴ
pi — The small circle changes ひ (hi) to ぴ (pi). Two strokes would make び (bi).
きゃ
kya — Small ゃ combines with き to make kya. Full-size きや would be ki-ya.
しゅ
shu — Small ゅ gives shu here. Compare しゃ (sha), しゅ (shu) and しょ (sho).
きって
kitte — Small っ marks a consonant hold: kitte, meaning postage stamp. It is not a separate tsu.
きょう
kyō — きょ makes kyo, and う extends the vowel in this word: kyō, today. The macron marks length.
みず
mizu — みず is mizu, water. Notice the dakuten on ず rather than reading the word as みす.
This practice checks recognition, not pronunciation. Results stay in this page session only.
How to read the hiragana chart
Read across the first row: a, i, u, e, o. The rows below combine those vowels with consonant sounds. Notice shi, chi, tsu and fu rather than trying to pronounce every row by mechanically adding the same consonant. The final ん has its own place in the chart.
Start with five characters. Cover the readings and point to them in a different order. If a shape disappears from memory, check it and return to it after another character. Add a new row when the previous group feels usable, not simply because you reached the end of a session.
Romaji is a starting aid, not a recording of Japanese pronunciation. Pair your reading with a reliable audio model. This reference shows printed forms; use a handwriting reference when learning stroke order.
Dakuten and handakuten: the marks change the sound
The two small strokes in が are called dakuten. The small circle in ぱ is called handakuten. Look for these marks as part of the character rather than recognising the large shape and skipping the rest.
The examples here introduce the changes; the 46-character chart above is the unmodified basic set. For a fuller reference, use the Japan Foundation chart linked below.
| Basic kana | With a mark | Reading |
|---|---|---|
| か | が | ka → ga |
| さ | ざ | sa → za |
| た | だ | ta → da |
| は | ば / ぱ | ha → ba / pa |
| し / ち | じ / ぢ | shi / chi → ji |
| す / つ | ず / づ | su / tsu → zu |
Small kana and long vowels need attention
A small ゃ, ゅ or ょ combines with a suitable preceding character: きゃ is kya, while full-size きや is ki-ya. Look at the size, not only the shape. In きょう, the small ょ belongs with き; the following う extends the vowel in this word.
A small っ holds or closes the following consonant sound, as in きって. Read it as part of the word rather than saying a separate tsu. Long vowels also belong to the spelling. Learn words such as おかあさん and せんせい with an audio model instead of applying one guessed spelling rule to every long sound.
きょう
kyō
today
きって
kitte
postage stamp
Read six first words
Try each word before looking at its romaji and meaning. These are shown in hiragana for practice; several are also commonly written with kanji. Choose two and find their characters in the chart in reverse order.
いえ
ie
house
ねこ
neko
cat
すし
sushi
sushi
みず
mizu
water
かぎ
kagi
key
はな
hana
flower (in this vocabulary example)
When should you add katakana?
You can begin a small katakana group while continuing to review hiragana. The sounds overlap, but recognition is a separate task. Compare one pair at a time, then practise the scripts in their own words. You do not need perfect recall of every hiragana before seeing コーヒー on a menu.
Three traps to catch before they become habits
First, do not let the chart position supply the answer. Recognising け after く is easier than meeting it alone. Use the reading checks above, then point to characters in a new order yourself.
Second, check character size. きや contains two full-size kana; きゃ uses small ゃ to make one contracted sound unit. With っ, size changes the job completely: small っ marks a consonant hold rather than a separate tsu.
Third, separate recognising a printed character from writing it. Handwritten き and さ may look less connected than their printed forms. Follow a reliable stroke model if your goal includes handwriting; copying the silhouette of a font can create awkward strokes.
きや / きゃ
ki-ya / kya
Full-size や and small ゃ change the reading.
つ / っ
tsu / consonant hold
The small form has a different job.
A ten-minute return to the chart
Spend two minutes recalling yesterday’s small group without looking. Use the next three minutes to repair the uncertain characters, checking the chart only after an attempt. Spend another three minutes on a few short words. Finish by choosing two shapes to recognise when you return tomorrow.
When the same pair keeps causing trouble, place the two forms next to each other and describe one visual difference. Then put each back into a word. A comparison helps you notice; another encounter checks whether you remember.
- Pick a small set you can actually revisit.
- Keep the word’s marks and vowel length in the recall attempt.
- Review a difficult pair again after another task, rather than staring at it continuously.
Work out a word without using its chart position
Take the word たまご, meaning egg. First locate た, ま and ご separately. The last form contains dakuten, so recognising こ alone is not enough. Say or identify the three parts, then read the complete word and connect it with its meaning. This small process shows how isolated character knowledge turns into a vocabulary reading.
Now shuffle the three characters and identify them again. Reading them correctly in the original word may partly depend on knowing what the word should be. A changed order checks the forms more directly. Return to たまご afterward so the exercise finishes with meaningful language rather than a random sequence.
For a second word, use さかな, meaning fish. Identify the characters, read the word and then compare the character shared with other vocabulary you know. Do not add many words at once. Two or three carefully inspected examples provide enough opportunities to notice whether a shape, mark or position is causing the hesitation.
Distinguish the three familiar particle pronunciations
The hiragana は is usually read ha as part of a word, but the topic particle は is pronounced wa. Similarly, へ is normally he in vocabulary but is pronounced e when used as the direction particle. を is conventionally pronounced o in the object-particle use shown in beginner sentences. Learn these as grammatical uses, not as reasons to stop trusting the chart.
Compare はな, “flower”, with 私は学生です。The same visible は has a different role in the sentence and therefore a different reading in these examples. The word はな does not become wana merely because you have learned the topic particle. The surrounding structure tells you which use you are seeing.
You do not need to identify every particle before beginning kana study. Meet a few clear models and return to the distinction when it appears. A word chart teaches ordinary character readings; a sentence lesson shows special grammatical uses. Both resources can be correct while supplying different kinds of information.
Read long vowels through whole-word spellings
Hiragana does not use one completely uniform spelling for every long vowel. Familiar words include おかあさん, せんせい and きょう. Learn the printed sequence together with a reliable recording. Applying one guessed spelling rule to every word can create errors that are difficult to detect from a romanised translation alone.
In きょう, small ょ combines with き, and the following う participates in the long vowel of this word. Full-size よ would create a different sequence. You need to attend both to character size and to the complete spelling. A vocabulary card that deletes the final う because the word already seems recognisable is incomplete.
When checking yourself, point to each written element and explain its role in the word. Then listen and read the complete form naturally. The aim is not to insert an exaggerated pause between every kana; it is to understand how the spelling represents the word you hear.
Learn to recognise a form in more than one visual setting
A printed character can look somewhat different across fonts, and handwriting introduces further variation. Begin with one clear model so the basic form is stable. Then compare a second reliable printed example or handwriting reference. Identify which features remain essential and which change with the style.
Characters such as き and さ may look less connected in ordinary handwriting than in some printed fonts. Do not assume that copying the outline of a typeface gives you the usual stroke movement. If handwriting is your goal, use a stroke-order model and compare your result with an appropriate handwritten example.
For recognition, change the setting modestly: a chart, a word list and a short sentence are enough. If a character becomes unrecognisable outside its familiar card, inspect the feature that changed. Repeated exposure should broaden recognition, not tie it more tightly to one exact image.
Keep a small diagnostic record for difficult forms
Create a note with three columns: what you saw, what you read and what the correct reading was. Add a short explanation only when it identifies the source of the error. “Missed the dakuten” is different from “confused the main shape”. This distinction tells you whether to practise marks, visual contrast or the word itself.
Review only a few entries at the start of the next session. Then read a fresh word containing the target form. If recognition improves, remove the item from the special list and let ordinary reading provide later encounters. The list should remain small enough to guide attention rather than become a permanent catalogue of every mistake.
A page of checked characters is not the final goal. Finish by reading something meaningful, however short. You are making hiragana available so that future attention can move toward vocabulary, grammar and the message of a sentence.
What changes between かき and かぎ?
Reveal the explanation +
かき is kaki; かぎ is kagi, meaning key. Check the small marks as well as the main shape. Recognising き alone is not enough to read が, ぎ and the other marked characters accurately.
Before you turn the page.
Are there 46 hiragana or more than 46?
There are 46 characters in the basic modern set shown here. Dakuten, handakuten and combinations such as きゃ extend what you can write. The additional references on this page show 25 marked forms and 33 common contracted combinations; they do not change the basic-set count.
Why do じ and ぢ have the same romaji here?
They normally share the pronunciation represented as ji in standard Japanese, while remaining different spellings. Likewise, ず and づ are shown as zu. Learn the spelling of the complete word instead of choosing a character only from the sound. For keyboard input, ぢ is commonly entered as di and づ as du.
When can I stop using romaji?
Try hiding it for a small familiar group as soon as you can make an attempt. Keep it available for checking, then gradually read more words without it. You do not need to remove all support on the same day, and romaji should not replace listening to a pronunciation model.
Do I need perfect hiragana before reading sentences?
No. A short supported sentence can give the characters a purpose while you continue reviewing them. Choose a text with only a few unfamiliar forms, use furigana where helpful and come back to the meaning after each lookup.
Why is は read differently in a sentence and in はな?
In はな it is part of the word and read ha. As the topic particle, は is pronounced wa. The written form is the same, but its grammatical role differs. Learn a clear sentence example alongside an ordinary vocabulary example so the two uses do not compete as unexplained exceptions.
Should I write every character repeatedly until it looks perfect?
Only practise handwriting to the degree your goal needs it. Use an accurate model, make a few attentive attempts and compare one from memory. Keep time for reading the forms in words. Copying a page can improve a movement, but it does not automatically demonstrate independent character recognition or comprehension.