If English spelling has ever made you despair — though, through, tough, thought, thorough, all sharing the same four letters and not one shared sound between them — Czech is about to feel like a holiday. Czech orthography is phonemic: it was designed, deliberately and relatively recently, so that the way a word is written tells you almost exactly how it sounds. There is roughly one letter for one sound. Silent letters barely exist. Once you know what each letter and diacritic represents, you can read aloud a Czech word you have never seen before and a native speaker will understand you.
This is one of the most encouraging facts about learning Czech, and it pays off enormously early on. In English, every new word is a small gamble — you cannot be sure how it sounds until you have heard it. In Czech, the spelling is the pronunciation, with only a short list of systematic adjustments to layer on top. The goal of this page is to give you that list, so you know exactly where the small gaps between letter and sound are — and trust that everywhere else, you simply read what you see.
What "phonemic" buys you
Take the word kalendář (calendar). There is nothing to decode and nothing to guess. You read it straight off: k-a-l-e-n-d-á-ř. The á is simply a long a (held about twice as long), the ř is the famous Czech sound (covered on the ř page), and every other letter has its plain value. No silent e, no surprise that a is actually pronounced like ay somewhere, no doubled letters that collapse into one sound. What you see is what you say.
Máš nový kalendář na příští rok?
Do you have a new calendar for next year?
The same is true of most of the language. Long words that look terrifying — nejnebezpečnější (the most dangerous) — are still read syllable by syllable exactly as written, with stress predictably on the first syllable. There is no second layer of "but actually it's pronounced...".
The five overlay rules
Here is the entire list of systematic gaps between Czech spelling and Czech sound. Master these and you have covered essentially all of them.
1. i and y sound identical
Modern Czech pronounces i and y (and their long versions í and ý) exactly the same — both are an ee-type vowel. The two spellings survive for historical and grammatical reasons, not for sound. This means být (to be) and a hypothetical byt-with-i would sound the same; the spelling distinction is something you learn as a writing rule, never something you hear.
Chci být doma do sedmi.
I want to be home by seven.
To víno bylo opravdu dobré.
That wine was really good.
For reading aloud, this rule is liberating: whenever you see i, í, y, ý, you make the same ee sound. You only have to worry about i-versus-y when you write, never when you read.
2. ě changes the preceding consonant
The letter ě (e with a háček) is never a sound on its own — it is an instruction to the consonant before it. After d, t, n it softens that consonant (so dě, tě, ně sound like ďe, ťe, ňe); after b, p, v it inserts a faint y-glide (so bě, pě, vě sound like bje, pje, vje); and mě is pronounced mňe. The classic example is děti (children), where dě is the soft ďe, not a hard de.
Naše děti chodí do školy pěšky.
Our children walk to school.
Město je dnes plné lidí.
The town is full of people today.
This is fully systematic — there is a dedicated page on the letter ě and what it does — so once you internalise it, ě is no longer a surprise but a reliable signal.
3. di, ti, ni are soft
Parallel to ě, the bare vowel i (or í) after d, t, n also softens those consonants. So di, ti, ni are pronounced ďi, ťi, ňi — the same soft consonants you would write explicitly as ď, ť, ň before other vowels. In ti of a word like dítě (child), the t is soft. By contrast, dy, ty, ny (with y) stay hard. This is the one place where the i/y spelling difference does change the pronunciation — not of the vowel, but of the consonant in front of it.
Ti studenti tady čekají dlouho.
Those students have been waiting here a long time.
Dítě usnulo hned po obědě.
The child fell asleep right after lunch.
See the soft consonants page for the full picture of how ď, ť, ň relate to di/ti/ni and dě/tě/ně.
4. Voicing assimilation and final devoicing
This is the overlay rule that reaches deepest into pronunciation. A voiced consonant at the end of a word is pronounced as its voiceless partner, and inside consonant clusters voicing spreads so that neighbouring consonants match. The cleanest example is led (ice), which is written with a final d but pronounced [let] — the d devoices to a t. Likewise hrad (castle) ends in a t sound, and lev (lion) ends in an f sound.
Na rybníku je silný led.
There's thick ice on the pond.
Ten hrad stojí na kopci nad městem.
The castle stands on a hill above the town.
Crucially, this is still predictable — it is a rule, not chaos. The spelling keeps the underlying voiced letter (which reappears when you add an ending: led but ledu with the d sound restored), and the rule tells you when to devoice. The full set of pairs and cluster behaviour is on the voicing assimilation page.
5. ú and ů are both [uː]
Czech has two ways to write a long u: ú (u with a čárka) and ů (u with a kroužek, the little ring). They sound identical — both are a long oo. The difference is purely about position and word history: ú appears at the start of words and after prefixes (úkol, task), while ů appears inside and at the end of words (dům, house; stůl, table). When reading, treat them as one sound.
Musím dokončit ten úkol do zítřka.
I have to finish that task by tomorrow.
Náš dům má modrá okna.
Our house has blue windows.
The writing rule for choosing between them is covered on the ů versus ú page. For pronunciation, there is nothing to choose — they are the same sound.
Putting the overlays together
Once you stop reading these adjustments as obstacles and start reading them as a short, reliable checklist, written Czech becomes transparent. Look at a sentence and the only questions are: any i/y (just say ee), any ě or di/ti/ni (soften the consonant), any word-final or clustered voiced consonant (devoice it), any ú/ů (long oo). Everything else is read at face value.
Včera večer jsme byli na koncertě v parku.
Last night we were at a concert in the park.
Ten obchod má otevřeno až do devíti.
That shop is open until nine.
In the first sentence, byli has the ee vowel (rule 1), koncertě has the soft tě (rule 2). In the second, obchod ends in a devoiced t sound (rule 4). Nothing else needs decoding. That is the whole machine.
Common mistakes
❌ Reading led as 'led' with a clear voiced d at the end.
Incorrect — a word-final d devoices to t.
✅ Reading led as [let].
Correct — final devoicing turns the d into a t sound; the d returns in forms like 'ledu'.
❌ Pronouncing děti with a hard de, as if spelled 'deti'.
Incorrect — ě softens the preceding d.
✅ Pronouncing děti as 'ďeti', with a soft d.
Correct — dě is the soft ď plus e.
❌ Assuming ú and ů are two different sounds and trying to hear a contrast.
Incorrect — there is no audible difference.
✅ Reading both ú and ů as a single long [uː].
Correct — they are spelled differently for historical reasons but sound the same.
❌ Hunting for silent letters and hidden English-style spellings in every new word.
Incorrect — over-thinking is the trap; Czech has almost no silent letters.
✅ Reading a new word like kalendář straight off, letter by letter.
Correct — outside the five overlay rules, what you see is what you say.
❌ Making a separate vowel sound for ě, as though it were a special e.
Incorrect — ě is an instruction to the consonant before it, not an independent vowel.
✅ Treating ě as 'soften the consonant, then say e (or je after b/p/v)'.
Correct — ě has no sound of its own.
Key takeaways
- Czech spelling is phonemic: outside a short list of rules, every letter has one reliable sound. Invest early in learning the alphabet and its diacritics.
- The five overlays are: i/y sound the same; ě softens or adds a glide; di/ti/ni are soft; voiced consonants devoice at word-end and assimilate in clusters; ú and ů are both a long oo.
- Each overlay is a predictable rule, not an exception. Learn the rules and reading aloud becomes automatic.
- The biggest English-speaker mistake is expecting English-style chaos and over-decoding simple words. Trust the spelling.
Now practice Czech
Reading grammar gets you part of the way. The exercises are where it sticks — free, no signup needed.
Start learning Czech→Related Topics
- The Czech Alphabet, háček and čárkaA1 — The 42-letter alphabet and the two diacritics that drive Czech spelling.
- Voicing Assimilation and Final DevoicingB1 — How consonants change voicing to match their neighbors and at word end.
- Soft Consonants: ď, ť, ň versus di/ti/niA2 — How the soft consonants ď, ť, ň are written — sometimes with a háček, sometimes hidden inside di/ti/ni and dě/tě/ně.
- The Letter ě and What It Does to ConsonantsA2 — How ě softens the preceding consonant and creates je/ňe sounds.
- ů versus ú: Two Ways to Write Long uA2 — When long u is spelled with a ring (ů) and when with an acute (ú).