Oznake odabrani model razumije — kliknite da bacite jedan u vaš tekst gdje se to događa:
Ovaj model čita običan tekst, tako da inline oznake se zanemaruju. Za tag-based emocije, prebaci na ekspresivni model kao što su Orfeus ili Bark.
Prirodno-jezični znakovi isporuke. Počašćeni od Qwen3-TTS danas; drugi ekspresivni modeli dolaze uskoro.
Ovaj model ne podržava upute o stilu — prebaci se na model koji radi (npr. Qwen3-TTS).
Definiši vlastite izgovore (riječ = izgovor):
-12
+12
Kontrola ekspresnosti (0 = neutralna, 1 = vrlo ekspresna)
Težina smjera bez klasifikacije (viša = više brza praćenja)
Opišite glasovni stil na prirodnom jeziku (parler koristi ovo umjesto predefiniranih glasova)
Dia dijaloški format: Koristi [S1] i [S2] oznake za različite zvučnike. Primjerice: [S1] Pozdrav tamo! [S2] Zdravo, kako ste?
Serbian is unusual in being fully digraphic: the same language is written in two parallel, fully interchangeable alphabets, a Cyrillic and a Latin one, so a robust text-to-speech engine has to read both scripts and map them to identical sounds. The orthography is famously phonemic, following Vuk Karadžić's principle of one letter per sound, which helps grapheme-to-phoneme conversion, but the synthesiser must still handle pitch-accent and vowel-length distinctions that change meaning. Serbian Latin uses digraphs such as nj, lj, and dž that correspond to single Cyrillic letters (њ, љ, џ), so the voice must treat them as single phonemes rather than letter sequences. Typical users include media and news outlets serving the Balkans, accessibility and e-government services in Serbia and Republika Srpska, localisation studios, and developers who need output in whichever script their audience prefers.
Uzorak — Српски / Srpski
“Вештачка интелигенција може претворити било који текст у природан српски глас за само неколико секунди.”
Ime
Српски / Srpski
Zvučnici
Roughly 9 to 12 million native speakers across the Western Balkans.
Obitelj jezika
South Slavic (Indo-European)
Skripta
Both Cyrillic and Latin (full digraphia)
Govoriš li?
Serbia, Bosnia and Herzegovina (Republika Srpska), Montenegro, and Serbian communities in Croatia, North Macedonia, and the diaspora.
Voicing news and media content for Serbian-speaking Balkan audiences
E-government and accessibility narration in both Cyrillic and Latin scripts
Audiobook and educational content production
IVR and call-centre voice prompts for regional businesses
Pronunciation aids for learners of Serbian
Serbian Tekst za govor — FAQ
Yes. Serbian is fully digraphic, and the synthesiser reads both the Cyrillic and Latin alphabets, producing identical pronunciation because the two scripts map to the same sounds.
They are treated as single phonemes, exactly as their single Cyrillic counterparts њ, љ and џ would be, so words are pronounced correctly rather than spelled out letter by letter.
The varieties are very close and mutually intelligible, but Serbian has its own standard pronunciation and the distinctive use of Cyrillic. The voice follows the standard Serbian norm.
Serbian has a pitch-accent system with rising and falling tones and vowel-length contrasts. The voice models natural prosody so output sounds fluent, though fine tonal minimal pairs may not be marked in plain text.
