Translate
Document translation that holds together
No cardFirst 5 pages freeYour work never trains AI
Literess, the AI editor in Transept
You: Make Grandma grumpier.
Literess: Done. She doesn’t just say it anymore. She grumbles it. sagte → raunzte
Ask in plain words. You approve every change. Meet Literess →
How does a file become a finished translation?
Drop the file in, then pick the language pair and a mode. Context, glossary, styleguide, and review then happen in one editor instead of across five tabs and a chat thread.
Upload or paste
Upload a DOCX, EPUB, Markdown, TXT, or PDF file, paste raw text, or import straight from Notion or Google Drive. Transept splits the file into blocks, and in a Word file or an EPUB each block goes back to its original position on export. A PDF, a scan or a photo of a page can be translated as text only or written onto its own pages.
Translate with document context
Each block is translated with the rest of the document in view: the chapter, the heading structure, the recurring terms. Your glossary goes into every request, so the terms on page 50 still match page 1.
Edit sentence by sentence
Click any sentence to see alternatives, regenerate it with an instruction, or rewrite it yourself. The wording you approve goes into your translation memory, and later runs get it as reference.
Export back to where you started
DOCX for clients, EPUB for e-readers, Markdown for engineers, Notion or Google Docs for your content team. A Word file, an EPUB, or a Google Doc goes back out in the layout it came in with.
Document translation is the hardest part of AI translation, and the place where generic tools break first. A DOCX with tables, footnotes, and a running header across 200 pages is more than a string of sentences: it’s a structured file in which the formatting carries meaning. A research paper in Word has citation markers that need to survive the round trip. A Markdown file from a documentation site has code blocks that must never be translated. Transept treats a Word file, an e-book, or a Markdown file as structured content from the moment you upload it. It splits the file into blocks that map back to their original positions on export, so the file that comes out reads like the file that went in, only in another language.
What is document translation built for?
Client deliverables
Freelance translators and agencies delivering DOCX files: a glossary per client, a styleguide per project, and QA before the email goes out.
Knowledge bases
Help centers, product docs, and policies, where every article in the library has to use the same terms.
Long-form content
Books, reports, white papers, and research: work where the voice has to survive the trip into another language.
10 languages, one weekend
We translated Transept itself into ten languages over three days, running our own pages through the same editor, glossary, and memory this site describes. If you are reading this in German, Japanese or Chinese, it came through that workflow.
Field notes from the weekend— Don’t take our word for it —
The editor, in miniature
A working slice of the real app: Literess, the glossary, the styleguide, workflows, and translation memory are all live. Click around.
Show how much of it is yours
Working with AI is not the same as letting it write for you. Transept keeps a record of every document: what the AI drafted, what you changed and what you checked. Download it as a PDF when a client, an editor or a publisher asks how the translation was made.
- Counted by word. The words the AI wrote, the words you changed and the paragraphs you wrote yourself, each counted on its own.
- Every step on record. Translations, rewrites, fixes, your edits and your approvals in one list, each line marked as yours or the AI's.
- Your sign-off, dated. Before the PDF downloads, Literess asks whether you read the translation through. Say yes and the report records it.
Two reports on the Free plan, unlimited on every paid plan.
Originality report
Chapter 3 · Ukrainian → English
Reviewed in full by the author, confirmed on Oct 21, 2026
Simple pricing
1 credit = 1 source word. Pro mode costs 3 credits per word. What each action costs
Free
Always freeFor your first projects
Unlimited users
Translated and proofread on Standard. About five with Pro.
- Pro mode: the model thinks each passage through, 3 credits per source word
- Standard quality: fast, accurate, faithful
- Translation memory: your past translations, reused to keep new ones consistent
- Glossary, styleguide, QA, import / export: all included
- Notion, Drive, TMX / XLIFF: import, export, and bring your own memory
- 68 languages: from Spanish to Tagalog
- Paragraph-by-paragraph translation in every mode
- Literess, your AI literary editor: 10 messages free, then credits
- Originality reports: 2 included
- No card required
Lite
The pocket edition
Unlimited users
- Whole pages and documents in one run
- Unlimited originality reports
- Workflows: chain translation, proofreading, and checks into one run
- API, MCP & webhooks: automate runs, or let AI assistants work in your account
- Teams: create as many as you need and invite anyone for free
- Top up anytime: pack credits never expire
Starter
Most popularFor regular client work
Unlimited users
- 3× Lite’s credits: for weekly work
- Teams: invite anyone for free; add €29/$33 seats to fund a shared credit pool (80,000 credits a month each)
Pro
For higher volumeFor agencies & large catalogs
Unlimited users
- 3× Starter’s credits: headroom for batch work and full catalogs
- Priority support: a direct line and faster replies
- Predictable bill: most months need no extra credit packs
FAQ
- We benchmark it on real documents, through the same pipeline you use (same prompts, same glossary and styleguide settings), and we read the outputs side by side. Quality depends on the language pair and the text, so the best test is your own: translate a page and read it against the source. The Free plan gives you about 11 pages to try. You choose the mode for each document, and every block shows which model wrote it.
- Upload DOCX, EPUB, PDF, Markdown, or TXT, import from Notion or Google Drive, or paste raw text. A Word file, an EPUB, or a Google Doc can go back out in its original layout, with only the text swapped. Any document also exports as a plain DOCX, PDF, Markdown, or HTML file, or to Notion or Google Docs. For other tools, export the translation data as TMX, XLIFF, or CSV.
- Yes for Word, EPUB, Markdown, and Google Docs: headings, lists, and bold and italic text all stay where they were. Transept splits the file into structured blocks instead of treating it as one wall of text, so the export comes back in the same shape. A PDF has its own choice: translate the text only, or write the translation onto the original pages (or onto rebuilt pages when the text runs longer).
- There is no hard limit. The Free plan includes 5,500 credits a month and every mode, enough to translate and proofread about 11 pages on Standard, or five with Pro mode; on Free you translate paragraph by paragraph. Translating a whole document in one run is on the paid plans. Lite, at €9/$10 a month, includes 25,000 credits, about 80 pages in Standard. Starter, at €29/$33 a month, includes 80,000 credits, which covers a book of about 265 pages.
- You can pick several target languages when you upload, or add them later, and batch translation runs the document into every language in one go. The same glossary and styleguide apply to each language, so your terms stay consistent.
- A chat window will translate the text of a DOCX, but the formatting is lost on the way out, and you rebuild headings, lists, tables, bold, and italics by hand. Transept keeps the structure from start to finish: the DOCX you export has the same headings, tables, and formatting as the original, in the new language.
- Transept reads a scan, or a photo of a page, with AI first (handwriting included), then translates it and can write the translation onto the original pages. A scan costs more than a digital page because every word is read from the image before it is translated, and the upload dialog shows each choice’s fee and time before anything is charged. If you still have the Word file the PDF was made from, that is the cheapest route: it keeps its formatting.
- Tracked changes are accepted before translation, and the resulting clean text is what gets translated. Comments in the source file aren’t translated, and the comments you leave in Transept stay in the workspace rather than in the exported file.
- There’s no length limit. The longest run we’ve timed and written up is a 40,000-word document, translated by several agents at once in about fifty minutes.
Keep reading around this capability
- Use caseTranslate your book with AI without losing the voice→A character glossary that holds for 300 pages, a styleguide built from your own chapter, and alternatives for every line of dialogue. For…
- FeatureA glossary the AI actually follows→Character names, product terms, brand phrases, client vocabulary. Build a glossary once (by hand, from a file, or from a sample document)…
- FeatureA styleguide that steers every regeneration→Tone, register, rules, formatting conventions. Write a styleguide yourself or have Transept generate one from a page of finished work. It…
Built for work you answer for
Private by default
Your text is never used to train AI, not by us and not by any provider we use, and it stays out of our logs. Delete your account and your text is deleted with it.
How we handle dataSecurity you can check
Hosted in Germany, in Hetzner’s ISO 27001-certified data center. Every sub-processor is listed, and a DPA is one email away.
All sub-processorsGive it context
Glossaries, styleguides, translation memory and per-string notes go with every request. You set them up once per project.
GlossariesYou’re the human in the loop
The AI drafts; you decide. Review every change side by side, keep or revert it, and send a review link when a client needs to look.
Review linksAPI, MCP and webhooks
The full API on every paid plan. Connect Claude or another AI assistant over MCP, or start runs from your own tools.
Developer guideSomething bigger?
Large catalogs, custom workflows, a DPA for procurement. Transept stays self-serve; write to us for the rest.
support@transept.aiDrop a 50-page document in and have it back, ready to deliver, before lunch
No card · First 5 pages free
Featured on
