Exact Cyrillic–Latin Transliteration

Serbian Cyrillic and Serbian Latin are designed for exact letter-by-letter conversion. Београд becomes Beograd, and converting Beograd back produces Београд. The word, grammar, and pronunciation do not change; only the graphic system changes. This is transliteration, not translation.

The crucial promise is one Serbian alphabet letter to one Serbian alphabet letter. Most partners occupy one character in both scripts, while Cyrillic љ, њ, џ correspond to the single Latin digraph letters lj, nj, dž. Exact work also preserves ј–j, ћ–ć, ђ–đ, ч–č, ж–ž, ш–š without English or Russian substitutions.

For practical decisions about changing alphabets within documents and conversations, continue with Switching Between Cyrillic and Latin.

The complete conversion chart

The chart follows Serbian Cyrillic dictionary order. Uppercase maps in the same way, subject to the special capitalization of Latin digraphs discussed below.

Work by letters, not by English sound guesses

Exact conversion does not ask, “How might an English speaker spell this sound?” It asks, “What is this Serbian letter’s partner?” Thus ј becomes j, not y; ш becomes š, not sh; ж becomes ž, not zh; and ч becomes č, not ch. English-oriented spellings are pronunciation aids at best and cannot be converted back reliably.

Russian conventions are equally inappropriate. Serbian uses ј, љ, њ, ћ, ђ, џ. It does not replace them with й, combinations containing ь, or Russian-oriented Latin sequences. For example, Serbian људи is ljudi, not lyudi, and њен is njen, not nyen.

The three digraph mappings

For transliteration, љ–lj, њ–nj, џ–dž each count as one mapping. A converter going from Cyrillic to Latin expands one character into two. A converter going from Latin to Cyrillic must inspect the complete digraph before converting individual characters.

Capitalization must also survive. Initial Cyrillic Љ, Њ, Џ become ordinary title-case Lj, Nj, Dž: Љиљана–Ljiljana, Његош–Njegoš, Џакарта–Džakarta. In all-capitals text they become LJ, NJ, DŽ: ЉИЉАНА–LJILJANA.

Rare boundaries that require word knowledge

One-to-one correspondence is exact at the level of Serbian letters, but a raw Latin character string can contain d + ž or n + j across a morphological boundary. Nadživeti “to outlive” is надживети, not наџивети, because it contains nad- + živeti. Injekcija is инјекција, not ињекција. The characters are adjacent, but they represent two Serbian letters in those words.

This is why “replace every nj with њ” is not a fully Serbian-aware algorithm. Reliable conversion uses a lexicon, morphological information, or exceptions. For ordinary native words such as њен–njen and поље–polje, the digraph mapping is straightforward.

Preserve more than letters

Exact transliteration normally preserves word division, punctuation, numbers, and capitalization. It does not modernize vocabulary, change Ekavian into Ijekavian, translate names, or “correct” the sentence into a different regional standard. Ekavian млеко–mleko remains Ekavian; Ijekavian Serbian млијеко–mlijeko remains Ijekavian. Changing mleko to mlijeko is dialectal or standard-variety conversion, not script conversion.

The second sentence is (Ijekavian Serbian); both sides of that pair retain the same reflex. Mixing млеко with mlijeko would silently change more than the script.

Structural equivalence and social distribution

Both alphabets can represent contemporary standard Serbian completely. They are structurally equivalent, but not socially interchangeable in every setting. Cyrillic is prominent and sometimes required in official or institutional contexts in Serbia; Latin is widespread in digital media, commercial contexts, and communication across the region. Follow an explicit script requirement and match the surrounding document when no requirement is stated.

That audience awareness does not change the mapping. A formal Cyrillic notice can be converted exactly into standard Latin, and an informal Latin message can be converted exactly into Cyrillic, provided both are spelled correctly. The register belongs to the wording; the script is a separate choice.

Conversion practice

Convert by identifying each Serbian letter, handling љ, њ, џ as units and preserving marks:

Common Mistakes

1. Converting by shape

2. Using English transliteration conventions

3. Stripping mandatory marks

4. Splitting an ordinary digraph

5. Treating Latin as merely informal

Key Takeaways

Map Serbian letters, not shapes or approximate English sounds. Preserve ј–j, љ–lj, њ–nj, ћ–ć, ђ–đ, ч–č, ж–ž, џ–dž, ш–š, along with capitalization, punctuation, and the original Ekavian or Ijekavian form. The two scripts are structurally equivalent; exact conversion and context-sensitive handling of rare boundaries make the result reversible.