The Japanese Alphabet: Hiragana, Katakana, and Kanji

Why the "Japanese alphabet" is really three scripts

If you search for the Japanese alphabet, you're really asking about a writing system that works differently from the one you grew up with. English has 26 letters, and you string them together to spell out the sounds of a word. Japanese doesn't do this. It writes with three separate scripts, and a single ordinary sentence usually contains all three woven together. Once that clicks, the whole system stops looking like chaos and starts looking like a set of tools, each doing a job the others don't.

Here are the three scripts in plain terms. Hiragana is a set of characters where each one stands for a sound, and it's used for native Japanese words and for the grammar that holds sentences together. Katakana is a second set of sound characters covering the exact same sounds, but it's used for words borrowed from other languages, for foreign names, and for a few other special jobs. Kanji is different in kind: these are characters adopted from Chinese, and instead of standing for a bare sound, each one carries a meaning and comes with one or more ways to read it.

Hiragana and katakana together are called kana. The word to know is syllabary, which just means a set of characters where each symbol is a syllable-sized sound rather than a single consonant or vowel. That's the key contrast with an alphabet. In English the letter "k" is a consonant with no vowel of its own. In Japanese kana there's no bare "k"; instead there are separate characters for "ka," "ki," "ku," "ke," and "ko." Each kana is a full little sound you can say on its own, which is part of why Japanese pronunciation is so even and rhythmic.

So why three scripts instead of one? History is the short answer. Japanese first borrowed Chinese characters, the kanji, to write a language that had no writing of its own. Over centuries, people simplified some of those characters into fast phonetic symbols, which became the two kana sets. The result is a system where kanji carries the heavy meaning, hiragana wraps around it to handle grammar, and katakana flags the words that came in from outside. It sounds complicated written down, but in practice each script signals something useful the moment you see it, and fluent readers lean on those signals constantly.

Hiragana: the base script (full chart)

Hiragana is where nearly everyone starts, and for good reason. It writes native Japanese words, it writes the grammatical endings on verbs and adjectives, and it writes the small connecting words that appear in every sentence. You cannot read even a beginner's textbook without it, so it earns its place as the very first thing to learn. The good news is that it's a closed set: there are 46 basic characters, and once you know them you can read any Japanese sound.

The traditional way to lay out the kana is the gojuon, which means "fifty sounds." It's a grid organized by consonant and vowel. Every Japanese vowel comes in five flavors, a, i, u, e, o, and the rows run through the consonants: k, s, t, n, h, m, y, r, w. Reading it is simple once you see the shape: pick a consonant row, pick a vowel column, and the character at the crossing is that combined sound. Here's the full basic hiragana chart with romaji, the roman-letter spelling of each sound.

A few things to notice, because they'll save you confusion later. The y row and w row have gaps, since sounds like "yi" and "wu" simply aren't part of the language. A handful of characters break the tidy pattern of their row: in the s row you get "shi" instead of "si," in the t row you get "chi" and "tsu" instead of "ti" and "tu," and in the h row you get "fu" instead of "hu." These aren't exceptions you need to fight; they're just how those sounds actually come out in Japanese, and the romaji reflects the real pronunciation.

The last character, ん (n), is the one lone consonant that can stand by itself with no vowel attached. It shows up at the end of syllables, like the n at the end of "Nihon," the Japanese word for Japan. It's worth flagging early because it's the only kana that behaves this way, and beginners sometimes forget it exists.

Voiced sounds: dakuten and handakuten

Once you have the basic chart, you get a big batch of extra sounds almost for free. Adding a small two-stroke mark called dakuten, which looks like a tiny quotation mark at the top right of a character, turns an unvoiced sound into a voiced one. So k becomes g, s becomes z, t becomes d, and h becomes b. A different mark, a small circle called handakuten, turns the h row into p sounds. You don't learn new shapes for any of these; you learn one base character and then recognize the mark on it.

You'll spot two sounds that appear twice: "ji" and "zu" each have two spellings. In modern Japanese the versions from the z row are the common ones, and the two from the d row show up in a small set of specific words. You don't need to agonize over the difference now. Just know that if you see the same romaji on two different characters, that's why, and the meaning of the word tells you which spelling applies.

Combined sounds: yoon

The last piece of hiragana is a set of glide sounds called yoon, made by taking a character from the i column, like ki, shi, or ni, and attaching a small ya, yu, or yo to it. The small character shrinks the two sounds into one smooth syllable. So ki plus a small ya becomes "kya," a single beat, not "ki-ya." These combinations cover sounds like the start of "Tokyo" and "Kyoto," so they're not rare.

The same trick works on the voiced characters too, giving you sounds like "gya," "ja," and "bya," but the pattern is identical, so once you see how the small ya, yu, and yo work, the rest follow. The important visual cue is size: a full-size ya, yu, or yo is its own separate sound, while a shrunken one glides onto the character before it.

Katakana: the script for foreign words (full chart)

Katakana is the second kana script, and here's the reassuring part: it covers the exact same sounds as hiragana, in the exact same gojuon layout. There are no new sounds to learn, only new shapes for sounds you already know. If you've learned hiragana, you've done all the hard conceptual work; katakana is a second coat of paint on the same wall. Most learners find it takes noticeably less time than hiragana did, precisely because the structure is already familiar.

What changes is the job. Katakana is used for words that came into Japanese from other languages, which is a huge and growing category. Coffee, computer, table, hotel, menu, and thousands of others are written in katakana, reshaped to fit Japanese sounds. It's also used for foreign personal names and place names, for the names of many animals and plants in casual or scientific writing, for onomatopoeia and sound effects, and sometimes for emphasis, the way English might use italics or all caps. When you see katakana, your brain should think "this is probably a borrowed or foreign word," and often you'll be able to guess it from the English underneath.

Here's the full basic katakana chart with romaji, laid out the same way as the hiragana chart so you can compare row by row.

Katakana takes dakuten and handakuten marks and forms yoon the very same way hiragana does, so カ ka with a mark is ガ ga, and キ ki plus a small ya is キャ kya. You don't need a separate lesson for any of it; the rules simply carry over. There's one small extra worth knowing: katakana marks long vowels with a straight line, ー, rather than by adding a vowel character. So the katakana for "coffee" ends up looking like ko, hi, long-mark, giving you "koohii."

A friendly warning about lookalikes. A few katakana are easy to confuse with each other, especially shi, tsu, so, and n. The difference is often the direction and angle of the strokes, and stroke order, which we'll cover shortly, is what keeps them straight in your mind and your hand. If you learn these four as a group and pay attention to how each is written, the confusion clears up fast.

Kanji: the characters that carry meaning

Kanji is the third script, and it works on a completely different principle from the two kana. A kana character stands for a sound. A kanji character stands for a meaning, and often a whole word. Take 日, which means "sun" or "day," or 山, which means "mountain," or 人, which means "person." You can often sense the origin: the character for mountain even looks a little like three peaks. Each of these packs a concept into a single symbol, which is why kanji does the meaning-heavy lifting in a sentence.

The wrinkle that trips up beginners is readings. Because kanji came from Chinese and got layered onto an existing spoken language, most kanji have more than one reading, and which one you use depends on the word. There are broadly two families of readings: one derived from the original Chinese pronunciation, used mostly when characters combine into compound words, and one native Japanese reading, used mostly when the character stands more or less alone. For example, 日 can be read "nichi" or "ni" in some words and "hi" or "bi" in others. You don't memorize a character in a vacuum; you learn it through the words it appears in, and the readings come along with those words.

Here's a small set of common kanji so the idea is concrete rather than abstract.

Notice how kanji combine. Put 日 (sun, origin) and 本 (book, origin) together and you get 日本, read "Nihon," the name of Japan itself, literally something like "origin of the sun." Compounds like this are everywhere, and they're part of what makes kanji efficient: two or three characters can express an idea that would take several kana to spell out phonetically, and a reader takes in the meaning at a glance without sounding it out.

How many kanji do you need? The set taught across Japanese schooling is roughly 2,000 characters, and knowing that range lets you read most everyday material like signs, menus, news, and books. That number scares people, but you never sit down to memorize 2,000 of anything at once. You learn a handful at a time, bound to real words you're using, and they accumulate over months and years the same way English speakers absorb thousands of words without ever studying a dictionary cover to cover. Japanese lands on many lists of the harder languages for English speakers largely because of kanji, and if you want the honest picture of that difficulty, our guide to the hardest languages to learn puts it in context.

How the three scripts work in one sentence

The payoff of learning all this is seeing how naturally the scripts share a sentence. Rather than competing, they split the work. Kanji handles the meaning-carrying words, hiragana handles the grammar that connects them and the endings that conjugate them, and katakana flags any foreign word in the mix. A skilled reader uses the visual difference between the scripts as a map, telling at a glance where one word ends and the next begins, which is genuinely helpful in a language that doesn't put spaces between words.

Take a simple example: 私は日本語をベンキョうします, which reads "watashi wa nihongo o benkyou shimasu," meaning "I study Japanese." Look at what each script is doing. 私 is a kanji meaning "I." は is a single hiragana, "wa," a grammar particle marking the topic. 日本語 is three kanji together meaning "Japanese language." を is another hiragana particle, marking the object. ベンキョう is written in katakana-free kanji and kana as the word for "study," and します is a hiragana ending that makes the verb polite. Every script pulls its own weight, and none of it is random.

This division is why you can't skip any one of the three and expect to read. Kanji without hiragana would be a pile of nouns with no glue. Hiragana without kanji is readable, and beginner material is often written entirely in kana, but it's slower to read and more ambiguous, since many words share the same sounds and only the kanji tells them apart. Katakana without the other two would leave you able to sound out loanwords and nothing else. The three-script system looks like extra work up front, and it is, but each part is solving a real problem, and together they make Japanese text unusually quick to scan once you're fluent.

How Japanese pronunciation works

One of the genuine gifts of Japanese is that pronunciation is regular and the sound inventory is small. Compared with English, there are far fewer vowel and consonant sounds to master, and once you know how each kana sounds, you know how to say any word written in kana, with no silent letters or surprise spellings. That's a real advantage for speaking, and it's worth understanding the handful of rules that shape how the sounds fit together.

Start with the five vowels, because everything is built on them. They're pure and consistent: a as in "father," i as in "machine," u close to the oo in "boot" but with the lips relaxed, e as in "bet," and o as in "or." Every kana ends in one of these five vowels except the lone n, and the vowels don't drift or diphthong the way English vowels love to. If you nail these five sounds cleanly and keep them steady, your Japanese will already sound far more natural, because learners who let their native vowels leak in are the ones who sound off.

The next idea is the mora, which is the beat Japanese counts in. Each basic kana is one mora, one even beat of time, and Japanese rhythm comes from giving every mora roughly equal length. A few things count as their own mora even though they attach to another sound: the lone n, a long vowel, and the small tsu that marks a doubled consonant. Take "Nippon," an emphatic reading of Japan's name: the small tsu in the middle creates a tiny pause that doubles the p, and it's a full beat. Getting this rhythm right, giving each mora its even beat rather than stressing syllables the English way, is one of the biggest levers for sounding fluent.

Long vowels matter more than beginners expect, because vowel length can change a word's meaning entirely. A short o and a long oo are different sounds in Japanese, and words that differ only in length are genuinely different words. Hiragana usually shows a long vowel by adding a vowel kana, while katakana uses that straight line, but either way you have to actually hold the long vowel longer when you speak. Rushing it is a classic mistake that makes speech sound clipped and can confuse a listener.

Finally, a word on pitch. Japanese uses pitch accent, meaning the melody of high and low across a word can distinguish meanings, unlike English which leans on loudness and length for stress. Pitch accent is real, and advanced speakers pay attention to it, but as a beginner you should not let it paralyze you. Get the vowels pure, give every mora its even beat, and hold your long vowels, and you'll be understood clearly. Pitch is a refinement you layer on later through listening and imitation, not a wall at the start. If you want to go deeper on producing clean sounds in any language, our guide on how to improve pronunciation lays out drills that transfer straight to Japanese.

Stroke order and reading basics

When you start writing kana and kanji by hand, you'll meet stroke order, the fixed sequence in which you draw the strokes of each character. It can feel like fussy tradition, but it earns its keep. Following the standard order makes your characters come out balanced and legible, it makes fast handwriting readable instead of a scribble, and it's what keeps those lookalike katakana, shi and tsu, so and n, distinct, because the order and direction of the strokes are exactly what tell them apart.

A few reliable rules cover most cases. You generally write from top to bottom and from left to right, the same broad direction as the whole writing system. Horizontal strokes usually come before vertical strokes that cross them. When a character has an enclosing box, you draw the outside frame before filling in what's inside, though the bottom of the box is often closed last. These aren't laws you memorize as a list so much as habits your hand picks up after writing a character a few times the correct way. Learn each character's order when you first meet it, and it sticks with almost no extra effort.

You don't strictly need to handwrite to read Japanese on a screen or to speak it, and plenty of modern learners type far more than they write. But writing characters by hand, even just a few times each, does something useful: it forces you to notice the exact shape and the parts a character is built from, which makes you a sharper reader and a faster recognizer. Even if your goal is mainly speaking, spending a little time writing the kana as you learn them tends to lock them in better than pure reading drills alone.

As for reading direction, Japanese is flexible. Modern books, websites, and most everyday text are written left to right in horizontal lines, exactly like English, so beginners rarely trip on this. Traditional printing, many novels, and newspapers also use a vertical style, running top to bottom in columns that read from right to left across the page. You'll grow comfortable with both through exposure, and you don't need to worry about the vertical style until you actually encounter it.

How to memorize hiragana and katakana

The two kana are a finite, closed set, which makes them one of the most satisfying things to learn in all of Japanese: you can genuinely finish them, and the finish line is close. The trick is to use methods that build fast recognition and recall rather than vague familiarity. Here's an approach that works for most people, and it's mostly about doing the right small things daily instead of one big push.

• Learn in small batches by row. Take the gojuon a row or two at a time, say five to ten characters, rather than trying to swallow all 46 at once. The vowel row, then the k row, then the s row. Small groups give your memory something it can actually hold, and finishing a batch feels like progress, which keeps you going.
• Use active recall, not passive staring. Looking at a chart over and over feels productive but teaches little, because you're recognizing, not retrieving. Instead, quiz yourself: see the romaji, produce the kana from memory, then check. See the kana, say the sound out loud, then check. The effort of pulling the answer up from nothing is what makes it stick.
• Say every character out loud. Don't recall kana silently. Voice each sound as you drill it, so you're training your mouth and ear at the same time as your eyes. For a script whose entire purpose is to represent sounds, silent practice throws away half the value.
• Space your reviews. Short daily sessions beat one long marathon, because memory forms through repeated retrieval spread over days. Ten focused minutes each morning will carry you through both kana far more reliably than a single exhausting afternoon that fades by the weekend.
• Read real words as soon as you can. The moment you know a few rows, start reading actual short words made from them. Sounding out いぬ (inu, dog) or にほん (Nihon, Japan) turns dry symbols into meaning, and reading in context cements the characters far better than isolated flashcards ever will.

Mnemonics can speed the early stage. Many learners attach a little story or image to each shape, picturing き (ki) as a key, for instance, so the shape and sound arrive together. Invent your own where they help, drop them once a character is automatic, and don't feel you need a crutch for every single kana. The mnemonics are training wheels; the goal is instant recognition with no story in between, and you get there through repeated reading. These same principles, active recall, spacing, and real context, apply to every word you'll ever learn in Japanese, and our guide on how to memorize vocabulary goes deeper on making them stick.

Practicing the sounds out loud

Here's the gap that catches many Japanese learners by surprise. You can learn to read every kana, recognize hundreds of kanji, and understand grammar on the page, and still freeze when it's time to actually speak. Reading is recognition: a character is in front of you and you decode it. Speaking is production: you have to summon the right sounds and words from nothing, in real time, while a person waits for your reply. Those are different skills, and the writing system, useful as it is, trains only the first one.

The good news is that Japanese pronunciation is friendly to production once you commit to practicing out loud. Because the sound set is small and regular, there are no hidden traps waiting to embarrass you, just a handful of habits to build: keep the five vowels pure, give every mora its even beat, hold long vowels their full length, and let the small tsu create its little pause. None of that is hard in isolation. What's hard is doing it all automatically at conversation speed, and the only way to get there is to say Japanese out loud, a lot, until the sounds stop needing conscious thought.

This is exactly where reading practice quietly falls short and where speaking practice has to take over. When you read, you set your own pace and there's no pressure to produce a single sound. When you speak, you're on the clock, you have to choose words and arrange grammar and make the sounds all at once, and any weakness in your pronunciation shows up immediately. Learners who only ever read and drill often discover, the first time they open their mouths, that they can't produce clearly the very sounds they thought they knew. The fix isn't more reading; it's speaking, early and often, from your very first rows of kana.

A practical way to start is small and out loud from day one. As you learn each kana row, say the sounds aloud rather than in your head. As you pick up your first words, こんにちは (konnichiwa, hello), ありがとう (arigatou, thank you), say them at full voice, hold the long vowels, and get the rhythm right. Then build to short sentences and, as soon as you possibly can, real back-and-forth conversation, because nothing else forces production under real conditions. This is the whole reason SPAIKING exists: it gives you an AI partner to speak Japanese with on demand, so you can practice producing the sounds out loud, get corrected in the moment, and build the speaking half of the skill that reading alone will never give you.

A sensible order to learn it all

Faced with three scripts, thousands of kanji, and a new sound system, a lot of beginners freeze over where to even begin. The order below is the one that keeps momentum, gets you speaking early, and never asks you to memorize something you can't yet use. Treat it as a path, not a strict schedule, and adjust the pace to your own life.

1. Hiragana first. Learn to read all 46 basic characters, plus the voiced sounds and yoon, over a week or two of short daily sessions. Say every one out loud. This is the foundation everything else sits on, and beginner material is written mostly in it.
2. Katakana next. With hiragana solid, add katakana, which shares the same sounds and layout and so takes less time. Practice by reading loanwords, since so many are close to their English source and give you quick, satisfying wins.
3. Basic sounds and first words, out loud. As the kana come together, start saying real words and simple greetings aloud, nailing the vowels and the mora rhythm. Don't wait until you feel ready; speaking clumsily from the start is how you learn to speak at all.
4. Kanji, steadily and in context. Begin adding kanji a small batch at a time, always tied to words you're actually using rather than as isolated symbols. A slow, constant trickle over months beats any attempt to cram hundreds at once.
5. Grammar and conversation, together. Layer in basic grammar and, above all, real speaking practice as early as you can. The scripts and sounds are tools; conversation is where they turn into an actual ability to communicate.

Notice what this order protects you from. You never try to read kanji before you can read kana, you never memorize characters you have no words for, and you never postpone speaking until some imagined day when you're "ready." Japanese rewards patience with its writing and boldness with its speaking, and the learners who make steady progress are usually the ones who start talking while their kana are still fresh, rather than hiding in study for a year first.

How SPAIKING fits your Japanese practice

Learning the Japanese scripts is a reading and recognition job, and you can do a lot of it on your own with charts, drills, and daily practice. What none of that trains is the act of speaking, of producing Japanese sounds and words out loud, in real time, in a real exchange. That's the exact gap SPAIKING is built to close. You learn the kana and the words on your own, and SPAIKING gives you a place to turn them into spoken conversation whenever you want, without needing to find a patient human partner first.

Here's how it fits the path above. Once you can read a few rows of kana and say your first words, you open SPAIKING and actually talk, reaching for the sounds and words you've been learning. Because the AI responds, you get the two-way pressure a textbook can never give you: a reply you have to understand and answer on the spot, which is precisely what drags a word from "I can read it" to "I can say it." And because SPAIKING corrects your pronunciation as you go, you find out in the moment whether your vowels are pure, your mora even, and your long vowels actually long, the details that separate clear Japanese from the muddled kind.

The parts that make it fit a beginner's routine:

• It's always available. No scheduling a tutor, no waiting for a language partner. Finish a study session and speak your fresh kana and words out loud seconds later, while they're at their most recallable.
• It's judgment-free. You can fumble a sound ten times and the AI just keeps the conversation going, which is what finally gets nervous beginners to produce out loud instead of retreating into more silent reading.
• It corrects the sounds you can't hear yourself get wrong. Your own ear forgives your own mistakes. Live correction points you at the vowel or the rhythm you're missing, so you fix it early instead of baking it in.
• It builds the half of Japanese that reading skips. The scripts give you the sounds on paper; speaking makes them yours. Doing both together is what turns a reader of Japanese into a speaker of it.

Think of the writing system and the speaking as two halves of one skill. The scripts, hiragana, katakana, and kanji, let you read and give shape to the sounds. Speaking is the half that's harder to arrange and easier to avoid, and it's the half that actually lets you communicate. Put them together, learn the kana and then say Japanese out loud in real conversation from early on, and you sidestep the trap that leaves so many learners able to read a language they still can't speak. You can start on the pricing page or head back to the SPAIKING home page to see how it works, and begin turning the Japanese alphabet into real spoken Japanese.