Translate copy and see the localization choices a reviewer should check.
Paste a UI string, a transactional email, or marketing copy, choose a target language and tone, and optionally lock your brand and glossary terms. You get a native-sounding translation — not a word-for-word machine output — plus a short block of localization notes that flag the judgment calls: an idiom that was re-cast, a string that may now overflow a button, a formal or informal address decision, or a brand term kept verbatim. It runs in your Oriora account; the tool is free and each translation is billed to your wallet. Your text is sent for one translation and never stored.
Paste
Drop in a UI string, an email, or marketing copy.
Choose
Pick a target language and tone; optionally lock brand & glossary terms.
Translate
One tap → a native-sounding translation plus localization notes.
Review & copy
Check the flagged choices, then copy the result. Nothing is stored.
How is this different from a raw machine translation?
Alongside the translation you get localization notes — the specific choices a reviewer should check: an idiom that was re-cast, a string that may overflow, a formal or informal address decision, or a brand term kept verbatim. A raw MT API gives you the text but not where it made a call.
Will it keep my brand names and code intact?
Turn on "Preserve brand & glossary terms" and it keeps brand and product names, code identifiers, {curly placeholders}, %s-style tokens, and URLs exactly as written — and adds a note listing what it kept.
Where does my text go?
Nowhere persistent. The text you paste is sent for one translation and never stored — no history, no translation memory. The AI translation is billed to your Oriora wallet.
Does it pick formal or informal address for me?
It honours the tone you choose. For languages with a T–V distinction (tu/vous, Sie/du, usted/tú, keigo) it states which form it used in a formality note so you can confirm it fits your audience.
Available on
In beta — launching soon.
Solo founders, indie devs, and small marketing teams shipping into a new locale who need more than raw machine translation — they want to see where the translation made a call, without a human localizer on every string.