Languages

Written in the language. Not translated into it.

A translated sweeper reads like a translated sweeper. Copy is written natively in each language — local idiom, how your frequency is actually said out loud, and the legal wording your own regulator requires.

Numbers read the way people say them

"Ninety-nine one" in English, "quatre-vingt-dix-neuf virgule un" in French. A frequency written as digits and handed to a voice model gets this wrong almost every time.

Legal IDs per country

Each regulator words station identification differently and orders the parts differently. The legal template is built from your station record and locked — the writer can add around it but never inside it.

A voice roster per language

Each language has its own curated roster, its own pronunciation dictionary and its own speech-to-text check. A native voice usually sounds better than one voice stretched across everything.

Optional

Or keep your own voice, in every language

If you already have a voice your listeners know, we can clone it once with their written consent and then use it across every language we support — no session to book, no second talent to hire, and the same voice carrying your whole sound. Your talent earns a royalty on every cut it voices and can withdraw consent at any time.

Where each language stands

The full list is read from our voice vendor at runtime, so it grows as they add languages. This table is what we have reviewed ourselves.

Language Spoken imaging Sung jingles Site interface
English · English Standard Available Yes
French · Français Standard In testing Yes
Spanish · Español Standard In testing Yes
Arabic · العربيةRTL Beta In testing —
Dutch · Nederlands Beta In testing —
German · Deutsch Beta In testing —
Hebrew · עבריתRTL Beta In testing —
Hindi · हिन्दी Beta In testing —
Italian · Italiano Beta In testing —
Japanese · 日本語 Beta In testing —
Polish · Polski Beta In testing —
Portuguese · Português Beta In testing —
Standard languages have been verified by native speakers on our side and are sold with no caveat. Anything your audience speaks that is not listed is still available — it is simply labelled beta until we have verified it, and we would audition a beta-language cut carefully before putting it on air.
Sung jingles are narrower than spoken imaging: the music model handles fewer languages well than the voice model does. We test each one before offering jingles in it, and where it does not hold up we simply do not offer them in that language. Spoken imaging is unaffected.

Need a language that is not on the list?

Tell us which one. If our vendor supports it, we can usually have it tested within a week.

Ask about a language
Something went wrong. Reloading the page usually fixes it. Reload ×

Rejoining the server...

Rejoin failed... trying again in seconds.

Failed to rejoin.
Please retry or reload the page.

The session has been paused by the server.

Failed to resume the session.
Please retry or reload the page.