Where every decision becomes memory
— From literary fiction to localization at scale —
Where every decision becomes memory
Transept AI is the AI-powered translation workspace that remembers your decisions — every terminology and tone choice carries across documents, languages, and your team, with glossaries, styleguides, and review built in.
Free to begin · No card required
LiteressHere to help you translate
Chill, love
No project
EN → UK
\+ Add language
ProofreadReviewM
Here to help you translate
Block #2 / 1339
Original
Sofia had finally done it. After many semesters of sharing dorm rooms with strangers who left dirty dishes in the sink and played music at ungodly hours, she’d secured a single room. Her own space. Her own schedule. Her own blissful, uninterrupted privacy.
Translation
Софія нарешті цього домоглася. Після багатьох семестрів у гуртожитках, де доводилося ділити кімнату з незнайомцями, які вічно лишали брудний посуд у раковині й вмикали музику о несвітній порі, вона вибила собі одномісну кімнату. Власний простір. Власний графік. Власна блаженна приватність, яку ніхто не порушував.
Block #3 / 1339
Original
This lasted only one summer.
Translation
Це тривало лише одне літо.
Block #4 / 13392
Original
The knock came at 11:47 PM on September 1st, followed immediately by the sound of a key turning in her lock. Sofia shot up in bed, heart racing, as the door opened to reveal a girl dragging a massive suitcase behind her.
Translation
Стукіт пролунав о 23:47 першого вересня, і одразу ж почувся звук ключа, що повертався в замку. Софія підхопилася на ліжку, серце шалено калатало; двері відчинилися, і на порозі з’явилася дівчина, тягнучи за собою величезну валізу.
Block #5 / 1339
Original
“Oh! Sorry, sorry!” the stranger said in accented English, clearly as shocked as Sofia. “I think… this is my room? Number 811?”
Translation
«Ой! Вибачте, вибачте!» — промовила незнайомка англійською з акцентом, вочевидь, шокована не менше за Софію. — «Здається… це моя кімната? Номер 811?»
Memory
Import
BlocksCommentsChatDocuments
Search
Try:key in the lockкімнатуaccent
Search
Scope
Everywhere
Exclude this docInclude other language pairsEnhance results
AI context
Include discussionsInclude alternativesContext
Manage TM settings
Showing matches for the selected block
71%
A key turned in the lock and the door swung open.
У замку повернувся ключ, і двері розчахнулися.
Chill, love — chapter 2you · last week
This doc55%
The knock came just before midnight.
Стукіт пролунав перед самою північчю.
Chill, loveyou · today
Alerts2ChatGlossaryStyleWorkflowsMemory
Auto-saved · just nowSearching everywhere you’ve translated
Interactive demo — click anything
Featured on
- I.For translatorsBatch translation, QA built in, and a styleguide per client — so a 50-page job reads like one translator wrote it.Translate faster, deliver better →
- II.For content creatorsDefine your tone once. Batch your backlog. Ship in any language before the publishing window closes.Publish in your voice →
- III.For localization teamsShared glossaries, shared styleguides, comment threads on every block — twelve languages from one source of truth.Scale your team workflow →
- I.
- II.
- III.
- IV.
- V.
- VI.
- ·Standard quality — fast, accurate, faithful
- ◆Translation memory — a database of past translations, reused for consistency
- ·Workflows — chain translate, proofread, and checks into one run
- ·Glossary, styleguide, QA, import / export — all included
- ·Notion, Drive, TMX / XLIFF — import, export, bring your own memory
- ·50+ languages — from Spanish to Tagalog, shelf to shelf
- ·10 lifetime messages with Literess — your agentic AI literary editor, who remembers your projects and preferences
- ·No credit card required Begin without paying →
- ✦Pro mode — deeper model for nuance, 3 words per source word
- ✦Batch translation across whole projects
- ✦Unlimited Literess — your agentic AI literary editor: drafts, proofreads, sets things up for you
- ✦Teams — invite anyone free; add €29 seats to fund a shared word pool (100,000/mo each)
- ✦Top up anytime — extra words never expire Start translating→
- ◇3× Starter’s allowance — headroom for batch work and full catalogs
- ◇Priority support — direct line, faster replies
- ◇Predictable bill — most months covered without top-ups Begin at scale→
- Words used50,000 of 100,000 included on Starter
- ### What's the difference between Standard and Pro mode?+ Standard is fast and accurate — great for most content. Pro mode uses a more capable model that better preserves nuance, tone, and voice; it costs 3× the words from your plan but is worth it for novels, nuanced marketing, or anything where the writing matters. You can switch per document or per sentence.
- ### What file formats can I bring in?+ Upload DOCX, PDF, Markdown, or TXT. Or import directly from Notion or Google Drive, or paste raw text. Export back to DOCX, Notion, Google Docs, or Markdown — your workflow stays yours.
- ### Is the free plan really free?+ Yes — 1,500 words per month, no credit card required. That's roughly 5 pages, with every feature unlocked: glossary, styleguide, QA, comments, sharing, all of it. Upgrade only when you outgrow it.
- ### Can I keep my terminology and tone consistent across documents?+ Yes — that's the core of Transept. Build a glossary for your terms (character names, branded phrases, technical jargon) and a styleguide for tone, style, and rules. Toggle them on per document and the same terminology and tone carry across every translation, page 1 to page 500.
- ### Does Literess remember between documents?+ Yes. Tell her you prefer British spelling, or that you're translating a sci-fi trilogy, and she'll bring it back next time. Up to 20 notes per account — review or clear them whenever.
- ### Can I share documents with my team or a reviewer?+
Who it's for✦
Three audiences, one workflow
The product✦
What it actually does, in six steps
Bring in any document, get a translation
DOCX, PDF, MD, TXT — or paste raw text. Side-by-side editor, your language pair on the way in. Translation is the starting point, not the destination.
Side-by-side translation
A glossary that travels with you
Build once, toggle on per project. Terminology stays consistent — page 1 to page 500, language 1 to language 12.
Glossary in context
A memory that remembers your decisions
Every translation you approve is remembered — the wording, and the reasoning behind it: the alternatives you rejected, the comments that explained the call. Transept hands that back to the model on the next run, across your documents and across your team. We call it decision-context memory — not a lookup you copy from, but memory that carries your judgment, not just your words.
Decision-context memory
Steer any sentence with a direction
Sentence-level control. Add a custom direction ("more formal," "keep the metaphor") and your glossary and styleguide feed in automatically — so the rewrite obeys your terminology and stays in your tone. The rest of the page stays put.
Direction + your styleguide
Literess reads, remembers, and acts
A literary editor at your side — she remembers your projects and preferences between documents, reads the translation, and flags where the tone has flattened or the meaning has drifted. And she acts: run a workflow, draft a document, move a job across your board — on your say-so. Not a chatbot bolted on; an editor who does the work.
Literess at work
Review and ship
A clean reading view for the final pass. Comments on any block; reply and resolve in place. Send a link to a reviewer or client — no DOCX-by-email.
Review modePricing✦
Simple pricing
Pay for words you actually translate — self-serve, no sales call to start. Top up when you need more; extra words don’t expire.
Standard mode spends 1 word per source word. Pro mode uses a deeper model — 3 words per source word. Manual edits, exports, comments, and review activity are free — words are only spent when a block is translated by AI.
Free
Always free
€0/ month
1,500 words / month
For your first projects
Unlimited users
\~5 pagesper month
Standard mode, 1 word per source word.
✦Most chosen✦
Starter
€29/ month
100,000 words / month
For regular client work
Unlimited users
Standard
\~330
pages / month
Pro mode
\~100
pages / month
and Starter adds
Pro
For higher volume
€79/ month
300,000 words / month
For agencies & large catalogs
Unlimited users
Standard
\~1,000
pages / month
Pro mode
\~330
pages / month
and Pro adds
Prices are in EUR. VAT is added at checkout where applicable — EU consumers see local VAT, while EU businesses with a valid VAT ID pay the net price under reverse charge. Top up any paid plan with word packs — 50,000 words for €19, 200,000 for €69, or 500,000 for €149 — and pack words never expire. Working as a team? Creating a team and inviting people is free; add seats (2 or more) to fund a shared word pool — each seat contributes its plan’s monthly words, on one invoice. Cancel anytime; 7-day money-back guarantee on your first subscription payment, no questions asked.
✦ Cost calculator ✦
Estimate the cost of your next project
Short story
\~5 pages
Free
Novella
\~100 pages
Starter
Full novel
\~330 pages
Starter
Standard
Fast, accurate translations
Pro mode
Deeper quality, preserves nuance
Words to translate
50,000words
\~167 pages
1,000100,000200,000
Monthly bill€29.00
Covered by your €29 / mo Starter plan
The origin story✦
Nothing could translate what we wrote — so we built what could
Two writers, one problem.
— Vitalii.I wrote 120,000 words of satirical romance fantasy — mages drowning in debt, a multiverse economy that makes Wall Street look ethical, and a love story tangled up in all of it.
— Mariia.I write semi-fictional stories about real people, built around how it felt. Most of the people I wrote about were foreigners. We met in English, spoke in English, lived those moments in English. So I wrote them that way too.
“But I sometimes wonder what version of them I missed by not writing them in their language.”Mariia
— Vitalii.The same flatness hit me from the other side. Every tool I tried preserved the words but killed the voice. Tone, rhythm, wit, emotional texture — nothing held. It all came back flatter than it went in.
— Both of us.So we did what we know how to do. We read the machine-translation papers. We benchmarked Claude, GPT, Gemini, and a handful of open models on the same 50-page chapter — judging accuracy, tone, and whether the satire still landed. None of them held all three; each failed in a different way. So we built the tool we wished had existed when we started.
FAQ
❧
Every decision, remembered
Free to begin · No card required