# AI XLIFF Translate — complete documentation

> Automated translation of WPML XLIFF exports for WordPress. Preserves brand voice, HTML, shortcodes and JSON-LD structured data; pay-as-you-go per page with no subscription.

> Source: https://aixlifftranslate.com/docs/ — generated 2026-09-09

---

# How it works

> What happens to a WPML export between the moment you upload it and the moment you download it back - parsing, classification, translation, checking and rebuilding.
> Source: https://aixlifftranslate.com/docs/how-it-works/

WPML exports your content as XLIFF: a standard XML file where each piece of text is a
**translation unit** with a source (your original) and an empty target (waiting for a
translation). We fill in the targets and hand the file back structurally unchanged, so WPML
accepts it on import.

The interesting part is everything that happens between those two moments.

## 1. Parse

The file is read as XML, not as text. That distinction matters: a WordPress page is full of
markup that *looks* like content - block comments, shortcodes, HTML attributes, structured
data - and a tool that treats the file as a string will happily translate a CSS class name or
a URL slug.

Parsing gives us the unit boundaries WPML defined, plus the source and target language from
the file's own header. We translate into the language the *file* asks for, not one you pick in
the interface, because WPML exports one file per target language.

## 2. Find the duplicates

Real websites repeat themselves. A call to action, a footer disclaimer, a product blurb reused
across twelve pages - an export routinely contains far fewer *distinct* strings than units.

Every unit is fingerprinted by its source text. Identical sources are grouped, one member of
each group is translated, and the result is applied to the rest. You pay for the work done,
not the units counted, which is where most of the saving on a typical export comes from.

## 3. Classify each unit

Not all text should be treated the same way, and some should not be translated at all. Each
unit is routed to one of four strategies.

- **Body content** - paragraphs, headings, descriptions. Translated in your brand voice, with
  room to breathe.
- **Metadata** - page titles, meta descriptions, alt text. Translated under SEO constraints:
  these have length limits and a job to do, so they are handled more tightly than prose.
- **Structured data** - JSON-LD blocks describing your business, products or FAQs. Parsed as
  JSON, with only the human-readable fields translated. Identifiers, enumerations, URLs, dates
  and type names are left exactly as they were, because changing them breaks how search
  engines read your page.
- **Not translatable** - URLs, email addresses, slugs, brand names, template placeholders such
  as `%%title%%`. Passed through untouched.

Large HTML blocks get one more step. They are split into meaningful segments - an FAQ item, a
section - translated segment by segment, then reassembled into the original document
structure. Translating a 4,000-word HTML blob in one shot is how you lose a closing tag.

## 4. Translate

Each unit goes to the model with instructions assembled for that unit: your brand voice, the
protected terms that apply to it, what kind of content it is, and where it sits on the page.

Two things are held out of the model's reach entirely. URLs inside links are swapped for
placeholders before the text is sent and restored afterwards, so a model cannot "improve" a
link target. Protected glossary terms are declared up front and enforced afterwards.

You choose a quality tier - **Medium** or **Pro** - rather than a specific model. Medium is
the sensible default for body content at volume. Pro costs more per page and earns it on copy
that carries weight. The tier is recorded against the job, so you can compare results.

## 5. Check the result

A translation that reads beautifully but drops a `<strong>` tag is a broken page. Before
anything is accepted it is checked against its source for:

- **Markup integrity** - the same tags, in the same shape. If the model invents a tag the
  source never had, that tag is removed rather than the whole translation being discarded.
- **Shortcodes and placeholders** - `[gallery]`, `%%sitename%%` and friends must survive
  intact and in the same quantity.
- **Protected terms** - your brand names must still be present, spelled the way you spell them.
- **Length** - a translation several times longer than its source is a hallucination, not a
  translation.
- **Unchanged output** - output identical to the input usually means a failure. Not always,
  though: a proper noun legitimately translates to itself, so short identities and
  model-confirmed ones are accepted rather than fought.

A failed check triggers one stricter retry. If that also fails, the unit is **flagged for your
review** rather than quietly shipped. Nothing is presented as translated unless it passed.

## 6. Rebuild

Translated units are written back into a copy of your original file. The XML structure,
attribute order and unit ids are preserved exactly, because WPML validates them on import.

Every download is rebuilt from the stored translations at the moment you request it. Edit a
translation in the review screen and your next download contains that edit - there is no
cached archive that can drift from what you see.

## What you get back

XLIFF files with the same names and structure you uploaded, each marked with a suffix so you
can tell them apart from your originals, zipped together. Import them into WPML exactly as you
would import work from any translation vendor.

## What we deliberately do not do

- **We do not touch your database.** Translation happens on files. Nothing is installed in
  WordPress and nothing changes on your site until *you* import.
- **We do not hold your translations hostage.** They are standard XLIFF files. Stop using us
  tomorrow and everything you already translated stays in your WordPress install.
- **We do not machine-translate blindly.** Any unit that could not be verified is flagged,
  because a translation you cannot trust is worse than one you know to check.

---

# Why the second run costs less

> Duplicate detection and translation caching - what each one does, when it saves you money, and what makes a cached translation stop being reused.
> Source: https://aixlifftranslate.com/docs/caching-and-dedup/

Two separate mechanisms reduce what you pay. They are easy to confuse, because both amount to
"we did not call the model for that one", but they work at different scales.

## Duplicate detection, within one upload

When a batch is parsed, every unit is fingerprinted by its source text. Identical sources form
a group, one member is translated, and the result is applied to the whole group.

This is what makes a first run cheaper than its unit count suggests. Sites repeat themselves
far more than people expect: navigation labels, calls to action, product boilerplate, the same
footer sentence on every page.

Matching is exact. "Book now" and "Book Now" are two different strings and cost two
translations.

## Caching, across runs

Every translation is stored, keyed by the source text, the target language and the kind of
content it is. Upload the same page again next month and its unchanged paragraphs are answered
from that store instead of being translated again.

The practical shape: a first run of a site is close to a full-price run. Re-running it after
editing three paragraphs bills you for roughly three paragraphs.

### What the cache belongs to

Cached translations are scoped to a **project**. Each project is a distinct website with its
own voice and its own protected terms, so a translation cached for one project is never served
to another. Two projects translating the same sentence each get their own answer.

### What stops a cached translation being reused

- **The source text changed.** Any edit counts, punctuation included - it is a different string.
- **A different target language.**
- **The text is used differently.** The same words as a heading and as a meta description are
  cached separately, because they are translated under different constraints.
- **You asked for fresh translations.** Re-running a job and retranslating a single unit both
  bypass the cache on read, while still storing the new result for next time.

### What does not invalidate it

Changing your brand voice or your glossary does **not** retroactively re-translate what is
already cached. This is a deliberate trade: it keeps costs predictable and re-runs fast. If you
have made a significant voice change and want it applied to existing content, re-run the job
with fresh translations turned on and accept that you are paying for the work again.

## Seeing the effect before you commit

Before a batch starts you get an estimate with an upper and a lower bound. The lower bound
already accounts for the cache: it is what you pay if every predicted hit lands. The gap
between the two is, roughly, what your existing translations are saving you.

## What this means in practice

- Translate a site once, then keep it in sync cheaply.
- Do not split one website across several projects unless they really are different sites. You
  would be paying for the same translations twice.
- Small, frequent updates are the cheap path. Rewriting everything is the expensive one, and
  that is simply the cost of work that is genuinely new.

---

# Getting started

> From a WPML export to translated pages back in WordPress - the whole round trip, step by step, including what to check before you translate a hundred files.
> Source: https://aixlifftranslate.com/docs/getting-started/

The whole loop is: export XLIFF from WPML, upload it here, review, download, import back. No
plugin, no database changes, nothing installed in your WordPress.

## Before you start

You need WPML with Translation Management, and content already marked for translation in the
languages you want. Everything else happens here.

## 1. Try it without an account

Paste a paragraph or drop a single XLIFF file into the demo on the home page. You get real
output from the real engine - same classification, same checks - within a small free
allowance. It is the fastest way to see whether the quality suits your site before you commit
anything.

## 2. Create a project

A **project is one website.** It owns that site's brand voice, its protected terms, and its
translation cache.

Get this right at the start. Splitting a single site across two projects means paying twice
for the same translations and maintaining two copies of your glossary. Putting two genuinely
different sites in one project means they share a voice they should not share.

## 3. Set up a brand voice

A voice profile is what stops your translations reading like a machine. You can write one
yourself or point us at your website and have a draft generated from your existing copy, which
you then edit. See [Brand voice](/docs/voice-profiles/).

You cannot start a batch without one. That is deliberate - it is the difference between
translation and transliteration.

## 4. Export from WPML

In WordPress: **WPML - Translation Management**, select the content, choose the target
languages, and export. WPML produces XLIFF files, one per target language, and you download
them as a ZIP.

Keep the ZIP as WPML gave it to you. Do not rename the files or flatten the folders: the
filenames carry information used on the way back in.

## 5. Upload

Drop the ZIP here. Before anything is translated you get a preview:

- how many files and translation units it contains
- how many of those are unique after duplicate detection
- an estimated cost, as an upper and a lower bound
- which project it will land in, and which voice profile will be used

Nothing has been spent at this point. Confirm to start, or cancel and change something.

**Start small.** Run five or ten files first, read the output properly, and only then upload
the other two hundred. This is the single most useful habit with any translation tool.

## 6. Watch it run

Progress is per unit, not per file, so a large file does not look stalled. You can leave the
page - the work continues - and come back later.

If a run stops short, for any reason, you can continue it. It picks up exactly where it left
off and does not re-bill anything already translated.

## 7. Review

Every unit is shown next to its source. Two things are worth your attention:

- **Anything flagged.** These are units that failed a check, or fell back to the source text.
  They are flagged precisely so you do not have to read everything.
- **Your most visible pages.** Home, top landing pages, anything with a call to action.

You can edit any translation inline, mark units as reviewed, or ask for a single unit to be
translated again. Edits are saved immediately and are included in the next download.

## 8. Download and import

Download the ZIP. Every file is rebuilt from the current stored translations, so it reflects
every edit you just made.

Back in WordPress: **WPML - Translation Management - Import**. Import in modest batches rather
than all at once; WPML is happier that way and a rejected file is easier to spot.

Your translations now live in your WordPress install, as ordinary WPML translations.

## A sensible first project

1. Export five representative pages - one with heavy formatting, one with structured data, one
   plain article.
2. Set up your voice profile and add your brand names as protected terms.
3. Translate those five, review them properly, import them, look at them on the site.
4. Adjust the voice or the glossary based on what you saw.
5. Then run the rest.

Step 4 is the one people skip, and it is the one that determines whether the other 200 files
come out right.

---

# Brand voice

> What a voice profile is, how it changes the output, how to have one drafted from your own website, and how protected terms keep your names intact.
> Source: https://aixlifftranslate.com/docs/voice-profiles/

Generic machine translation produces text that is *correct* and sounds like nobody. It flattens
a friendly second-person site into corporate third-person prose, drops your contractions, and
turns your product names into descriptions of your product names.

A **voice profile** is the instruction set that stops that happening. It travels with every
unit we translate for your project.

## What is in a profile

**Brand identity.** Who you are, who you are talking to, and what you sound like. This is prose,
not settings - the more specific it is, the more useful. "Casual, like a knowledgeable local
friend, for travellers aged 18 to 35" beats "friendly and professional", which describes almost
every company that has ever existed.

**Per-language guidance.** Tone does not translate one-to-one. German has a formal and an
informal register and you must pick one. Italian marketing copy tolerates warmth that would read
as unprofessional in Dutch. Each target language gets its own notes.

**Rules for body content and for metadata separately.** A page title has 60 characters and an
SEO job; a paragraph has room to breathe. They should not be written to the same brief.

**Protected terms** - your glossary. See below.

## What you write, and what we handle

You write the *content* of your voice: identity, tone, rules, the things only you know.

The *structure* around it - the output contract, the formatting rules, the instructions that
keep markup and shortcodes intact - is ours, and it is applied consistently to every profile.
That split matters: when we improve how translations are constrained, every project gets the
improvement without anyone editing their profile.

You will not find placeholder tokens or fencing syntax in a profile, and they are rejected if
you try to add them. Write prose.

## Generating a profile from your website

Writing a good voice brief from a blank page is hard. You can instead give us your website URL
and pick your target languages, and get a complete draft back: brand identity, per-language
guidance, and a glossary of terms found on your site.

How it works:

1. We discover and read a sample of your pages.
2. Those pages are analysed to produce a voice description and a candidate glossary.
3. You review the draft and apply it, in whole or in part.

It is a **starting point, not an answer.** Read it before applying it. It will usually get your
tone roughly right and will sometimes propose glossary terms you do not want frozen. Editing a
good draft is far quicker than writing from nothing.

There is a per-day limit on generations, and you are shown a cost estimate before one runs.

## Protected terms (the glossary)

A glossary term is a word or phrase that must survive translation unchanged. Your company name.
Your product names. Place names you use in a particular form.

Without this, "Nest Pass" becomes "Nido Pase" in Spanish, and it does so *convincingly* - the
translation is grammatically perfect and completely wrong.

Protected terms are handled twice: declared to the model before translation, and verified
afterwards. A translation that lost a protected term does not silently ship.

Two details worth knowing:

- **Protection applies to visible text.** A term appearing inside a CSS class name or a URL is
  left alone - rewriting it there would break your page rather than protect your brand.
- **Import your existing list.** If you already maintain a glossary in WPML, you can import it
  as a CSV rather than retyping it.

## Getting a voice right

Start from real copy. Take two paragraphs from your site that sound exactly like you, and
describe *why* they sound like you - that description is your profile.

Then translate five representative pages and read the output in a language you actually speak.
Adjust, and only then run the rest. A voice profile improved after five pages is worth far more
than one theorised before the first.

Changing your voice does not retroactively re-translate existing content - see
[Why the second run costs less](/docs/caching-and-dedup/) for why, and what to do when you want
it applied anyway.

---

# Glossary of terms

> Every term this documentation uses - XLIFF, translation unit, project, voice profile, tier, cache hit, flagged unit - defined in one place.
> Source: https://aixlifftranslate.com/docs/glossary-of-terms/

Definitions for the vocabulary used across these pages and in the app.

## Content and file formats

**XLIFF**
: XML Localisation Interchange File Format - the standard exchange format for translation work.
  WPML exports it, and it is what you upload here. The version WPML produces is XLIFF 1.2, and
  that is the version this tool is built and tested against.

**Translation unit**
: One translatable piece of a file: a paragraph, a heading, a meta description, a structured
  data block. WPML decides the boundaries; we do not re-cut them. A unit is the thing that gets
  translated, counted and billed.

**Source and target**
: The source is your original text. The target is the translation. In a fresh WPML export every
  target is empty; filling them in is the job.

**Page**
: The billing unit in pricing. Pages are a normalised measure of volume, which is why a very
  long page and a very short one do not cost the same.

## Organising your work

**Project**
: One website. It owns that site's voice profile, its protected terms, and its translation
  cache. Translations are never shared between projects, because two projects are two different
  sites with two different voices.

**Job (or batch)**
: One upload, processed as a unit. A job contains files, files contain units. Job status,
  progress and cost are all reported at this level.

**Voice profile**
: The brand-voice instructions applied to every unit in a project - identity, tone, per-language
  guidance, and rules for body content and metadata. See [Brand voice](/docs/voice-profiles/).

**Glossary term (protected term)**
: A word or phrase that must survive translation unchanged - typically a brand or product name.
  Declared to the model before translation and verified afterwards.

## How text is handled

**Content type**
: The strategy a unit is routed to: body content, metadata, structured data, or not
  translatable. It determines the instructions, the constraints and the checks applied. See
  [How it works](/docs/how-it-works/).

**Structured data (JSON-LD)**
: Machine-readable markup describing your business, products or FAQs for search engines. Parsed
  as JSON so only human-readable fields are translated; identifiers, URLs, dates and type names
  are left exactly as they are.

**Placeholder**
: A template token such as `%%title%%` or `%%sitename%%` that WordPress or an SEO plugin
  replaces at render time. Never translated, and checked for after translation - a lost
  placeholder produces a visibly broken page.

**Shortcode**
: A WordPress `[bracketed]` tag. Preserved exactly, in the same quantity, in the same order.

## Cost and quality

**Duplicate**
: A unit whose source text is identical to another unit's in the same upload. Translated once
  and applied to all of them.

**Cache hit**
: A unit answered from a previously stored translation instead of being translated again.
  Requires the same source text, target language and content type, within the same project. See
  [Why the second run costs less](/docs/caching-and-dedup/).

**Fresh translations**
: An option that ignores stored translations on read while still storing the new results. Used
  when you want a genuinely new answer - after a significant voice change, for example.

**Tier**
: The quality level chosen for a job - **Medium** or **Pro**. Medium is the sensible default for
  body content at volume; Pro costs more per page and is worth it for copy that carries weight.
  You choose a tier rather than a specific model, so improvements reach you without you having
  to track which engine is behind it.

**Flagged unit**
: A unit that failed its post-translation checks, or that fell back to its source text. Flagged
  so it surfaces in review rather than shipping unnoticed. A flag means "look at this", not
  "this is broken".

**Review**
: Reading translations against their sources, editing what needs editing, and marking units as
  checked. Review state is per unit and is kept with the job.

## Round trip

**Import (into WPML)**
: Loading finished XLIFF back into WordPress through WPML's Translation Management screen. This
  is when your site actually changes - nothing before it touches your database.

**Reverse import**
: Loading *already translated* XLIFF into this tool - translations done elsewhere, or by a human
  - so they become reviewable and reusable here. It performs no translation and costs nothing.

---

# Frequently asked questions

> Straight answers on languages, file formats, quality, cost, ownership of your translations, and what happens to your content.
> Source: https://aixlifftranslate.com/docs/faq/

## About the product

### Do I still need WPML?

Yes. WPML manages multilingual content in WordPress: the language switcher, the URL structure,
which post is a translation of which. We do not replace any of that. We do the part WPML asks a
translator to do, and hand the result back in the format WPML expects.

### Do I have to install a plugin?

No. Nothing is installed in WordPress and nothing runs on your server. You export files, we
translate files, you import files. Your site is unchanged until you import.

### What happens to my translations if I stop paying?

Nothing. They are already in your WordPress install as ordinary WPML translations, and the
XLIFF files are yours to keep. There is no subscription to lapse and nothing that expires or is
withdrawn. This is the main structural difference from proxy-based translation services, where
cancelling can take your translated site with it.

### Which languages are supported?

Translation targets are driven by your WPML export, and the engine covers the major European
languages. The free demo is limited to a smaller set - English, Spanish, German, French and
Italian - because it runs synchronously in your browser.

The source language is read from the file itself. WPML always records it, so you do not choose
it.

### What file formats do you support?

XLIFF as exported by WPML, which is XLIFF 1.2. That includes content with Gutenberg blocks,
shortcodes, custom fields and custom post types - the formatting is preserved rather than
merely tolerated.

## Quality

### How good are the translations?

Good enough that the work becomes reviewing rather than rewriting, which is the honest bar. They
are not a substitute for a professional human translator on copy where a single sentence carries
real commercial weight.

The most useful thing you can do is [try the demo](/) on your own content. It runs the real
engine, so what you see is what you would get.

### How do I know a translation did not break my page?

Every translation is checked against its source before it is accepted: matching markup, intact
shortcodes and placeholders, protected terms still present, plausible length. A unit that fails
is retried once under stricter instructions, and if it still fails it is **flagged for review**
rather than shipped quietly. See [How it works](/docs/how-it-works/).

### Will it keep my brand names intact?

If you tell it which ones they are. Add them as protected terms and they are declared before
translation and verified afterwards. Without a glossary, product names get translated - fluently
and wrongly. See [Brand voice](/docs/voice-profiles/).

### Can I edit a translation I do not like?

Yes, inline, at any point. Edits are stored immediately and every download is rebuilt from the
current stored translations, so what you download always matches what you see on screen. You can
also ask for a single unit to be translated again.

## Cost

### How does pricing work?

Pay as you go, per page, from EUR 0.02 - cheaper at volume. No subscription, no monthly minimum,
no credit packs to buy up front. You are billed for translation work actually performed, which
is why duplicates and cache hits cost you nothing.

### Why is my second run so much cheaper?

Duplicate detection within an upload, and caching across uploads. Re-translating a site after
editing three paragraphs bills you for roughly three paragraphs. See
[Why the second run costs less](/docs/caching-and-dedup/).

### Can I see the cost before committing?

Yes. Every upload shows an estimate - upper and lower bound - before anything runs, along with
the unit and duplicate counts behind it. Nothing is spent until you confirm.

### Can I use my own AI provider account?

Not currently. The per-page price includes the translation itself, so there is no separate API
bill and no keys to manage.

## Your content

### Do you keep my content?

Your uploads and translations are stored so that jobs can be resumed, reviewed, re-downloaded
and served from cache on later runs. That storage is what makes the second run cheap. Demo
content submitted without an account is held only briefly and then removed.

### Is my content used to train models?

No.

### Can two of my sites share translations?

No, and deliberately. Translations are scoped to a project, and a project is one website with
one voice. Sharing them across sites would mean one site's tone leaking into another's pages.

## Practical

### How long does a batch take?

Roughly a minute per file, running sequentially - a pace chosen to stay well inside rate limits
rather than to win a benchmark. A hundred-file batch is something you start and come back to.
Progress is reported per unit, and you can close the tab.

### What if a run stops halfway?

Continue it. It resumes exactly where it stopped and does not re-bill anything already
translated. A run that ended early for any reason - a stopped job, an interrupted batch, an
allowance running out - is picked up the same way.

### Can I bring in translations done elsewhere?

Yes. Import already-translated XLIFF and it becomes reviewable here, and reusable as cache on
later runs. It performs no translation and costs nothing.

### Something is not covered here

The rest of these docs go deeper: [How it works](/docs/how-it-works/) for the pipeline,
[Getting started](/docs/getting-started/) for the round trip, and
[Glossary of terms](/docs/glossary-of-terms/) if a word here was unfamiliar.

---

# Privacy Policy

> What we collect when you translate a file, why we keep it, who else handles it, and how to have it deleted.
> Source: https://aixlifftranslate.com/docs/privacy-policy/

AI XLIFF Translate translates WPML exports for WordPress sites. This policy explains what happens to
your information when you use it, in plain language.

Questions, requests or complaints: [aixlifftranslate@gmail.com](mailto:aixlifftranslate@gmail.com).

## What we collect

**Your account.** Your email address, and the sign-in records our authentication provider keeps so you
can log back in. We never see your password.

**The content you upload.** The XLIFF files you send us and the translations produced from them. These
are the pages, headings, meta descriptions and structured data of your own website.

**How you use the service.** Which files you translated, when, how many pages, what it cost, and
whether a run succeeded. This is what your bill is made of and what lets a stopped run be continued.

**The free demo, without an account.** A short-lived identifier stored in your browser so the demo can
recognise you across a page reload and apply its limits, together with a one-way fingerprint of your
network address. The fingerprint exists to stop the free tier being abused, and cannot be reversed
into your address.

We do not collect special categories of personal information, and we do not build advertising
profiles. There is no third-party analytics or advertising code on this site.

## Why we keep it

- **To run the service.** Translating a file and giving it back to you is the whole purpose.
- **To let you come back to your work.** Uploads and translations are stored so a job can be resumed,
  reviewed, re-downloaded, and served from cache on later runs. That storage is what makes a second
  run cheap.
- **To bill you** for work actually performed.
- **To keep the free demo free** by detecting abuse of it.
- **To meet our legal obligations,** such as keeping billing records.

## Who else handles it

Delivering a translation means sending your text to systems we do not own:

- **Third-party AI model providers,** which perform the translation itself. Your text is sent to them
  for that purpose and no other.
- **Our hosting, database and authentication providers,** which store and serve the application.

These are service providers acting on our instructions, not independent recipients free to do as they
like with your content. We do not sell your information, and we do not share it for advertising.

## Your content is not training material

Your content is not used to train translation models, ours or anyone else's. We do not grant our
providers the right to use it that way.

## How long we keep it

Your uploads, translations and job history stay for as long as your account is open, because that is
what makes them re-downloadable and re-usable. Delete your account and we delete them.

Demo content submitted without an account is held only briefly and then removed.

Billing records are kept for as long as accounting and tax rules require, even after an account
closes.

## Cookies

We use two, both strictly necessary and neither used for tracking:

- a **sign-in cookie**, so the application knows it is you between page loads;
- a **demo cookie**, so an anonymous demo run survives a reload and its free allowance can be counted.

There are no advertising or analytics cookies, so there is no consent banner to click away.

## Your rights

If you are in the UK or the European Economic Area, you have the right to ask for a copy of your
information, to have it corrected, to have it deleted, to restrict or object to how we use it, and to
receive it in a portable form. You may also complain to your national data protection authority.

You can exercise any of these by writing to
[aixlifftranslate@gmail.com](mailto:aixlifftranslate@gmail.com). We will respond within one month.
There is no charge.

## Where your information is held

Our providers may store and handle information outside the country you live in, including outside the
European Economic Area. Where that happens, transfers are covered by the safeguards required under
applicable data protection law.

## Security

Traffic to and from this site is encrypted. Access to stored content is restricted to the account it
belongs to, and translations are scoped to a single project, so one site's content is never served to
another.

No service can promise perfect security. If a breach ever affects your information, we will tell you
and the relevant authority as the law requires.

## Children

This is a tool for people running websites. It is not intended for children, and we do not knowingly
collect their information.

## Changes

If this policy changes materially we will update the date at the foot of this page and, where the
change affects you, tell account holders by email.

## Contact

[aixlifftranslate@gmail.com](mailto:aixlifftranslate@gmail.com)

---

# Terms of Service

> The agreement covering your account, your content, what the service does and does not promise, and how billing works.
> Source: https://aixlifftranslate.com/docs/terms-of-service/

These terms cover your use of AI XLIFF Translate. By creating an account or using the free demo, you
agree to them.

Questions: [aixlifftranslate@gmail.com](mailto:aixlifftranslate@gmail.com).

## The service

AI XLIFF Translate takes a WPML export from a WordPress site, translates it while protecting markup,
shortcodes, placeholders, structured data and the terms you have marked as protected, and gives you
back a file you can import.

Translation is performed by machine. It is checked automatically before it is accepted, but it is not
a human translation and it is not certified.

## Your account

You need a working email address, and you are responsible for what happens under your account. Tell
us promptly if you think someone else has access to it.

One person or organisation, one account. Do not share credentials.

## Your content

**Your content stays yours.** Uploading a file transfers no ownership to us.

You give us only the permission needed to run the service: to store your files and translations, to
send your text to the model providers that translate it, to keep translations so runs can be resumed,
reviewed, re-downloaded and re-used from cache, and to give the results back to you.

**The translations are yours to keep.** They are files you download and own. There is no subscription
that can lapse and take them away, and nothing here expires or is withdrawn if you stop using the
service.

You confirm you have the right to upload what you upload, and that translating it breaks no law and
no one else's rights.

## Acceptable use

Do not use the service to process content that is unlawful, or that you have no right to process. Do
not attempt to break, overload or work around the limits of the service, and do not use the free demo
to avoid paying for production work.

We may suspend an account that does these things. Where it is reasonable to do so, we will tell you
first.

## Billing

Pricing is per page, published on the [pricing section](/#pricing) of the home page.

- **Nothing is charged before you ask for the work.** Every upload shows an estimate first, and
  nothing is spent until you confirm it.
- **You are billed for translation actually performed.** Duplicate text within an upload, and text
  already translated on a previous run, cost you nothing.
- **No payment card is required** to open an account, to run the free demo, or to translate.
- There is no subscription and no monthly minimum.

Refunds are covered by the [Refund Policy](/docs/refund-policy/).

## What we do not promise

The service is provided as it is. In particular:

- **We do not warrant that a translation is accurate, complete or fit for a particular purpose.**
  Machine translation makes mistakes. Review before you publish, especially anything with legal,
  medical, financial or safety consequences. The review tools exist for this reason.
- We do not promise uninterrupted availability. The service depends on providers we do not control,
  and may be unavailable for maintenance or for reasons beyond us.
- We do not promise that any particular language, format or feature will remain available.

## Liability

Nothing in these terms limits liability that cannot lawfully be limited, including for death or
personal injury caused by negligence, or for fraud.

Subject to that, we are not liable for indirect or consequential loss, for lost profits, revenue,
goodwill or data, or for anything arising from your reliance on a translation you did not review. Our
total liability for any claim is limited to what you paid us in the twelve months before it arose.

If you are using the service as a business, you accept it is your responsibility to check translated
content before publishing it.

## Ending it

You may stop using the service and close your account at any time. Deleting your account deletes your
stored files and translations, so download anything you want to keep first.

We may end or suspend access if these terms are seriously or repeatedly broken. Amounts already owed
for work performed remain payable.

## Changes to these terms

We may update these terms. The date at the foot of this page shows when they last changed, and
account holders will be told by email about changes that materially affect them. Continuing to use
the service after a change means accepting it.

## Contact

[aixlifftranslate@gmail.com](mailto:aixlifftranslate@gmail.com)

---

# Refund Policy

> A seven-day refund on any translation, why you are never charged before the work happens, and how to ask for your money back.
> Source: https://aixlifftranslate.com/docs/refund-policy/

Short version: **you can have your money back within seven days, for any reason.** And because you are
only ever charged after work you asked for, there is usually nothing to refund in the first place.

## You are not charged up front

The billing model does most of the work a refund policy normally has to:

- **No payment card is required** to open an account or to translate.
- **There is no subscription,** no monthly minimum, and no credit pack to buy in advance.
- **Every upload shows an estimate before it runs** - an upper and a lower bound, with the page and
  duplicate counts behind it. Nothing is spent until you confirm.
- **You are billed for translation actually performed.** Duplicated text within an upload, and text
  already translated on an earlier run, cost nothing.

So you can see the price, decline it, and walk away having spent nothing. You can also try the
[free demo](/) on your own content first: it runs the real engine, so what you see is what you would
get.

## The seven-day refund

If you are unhappy with a translation, ask within **seven days** of being charged for it and we will
refund it.

- It applies to any translation work billed in that window.
- You do not have to justify it. "It is not what I wanted" is enough.
- You keep the files you have already downloaded. We are not going to ask for them back.

Write to [aixlifftranslate@gmail.com](mailto:aixlifftranslate@gmail.com) with the account email and
roughly which job you mean - the date and the file names are plenty. We will confirm by email, and
the money goes back the way it came.

## After seven days

The seven-day window is the guarantee. Past it, write to us anyway.

If the service billed you for something it did not do, double-charged you, or failed in a way that is
ours to own, we will put it right whatever the date. A policy is not a reason to keep money we should
not have taken.

## What a refund does not cover

Being honest about the limits, so nothing here reads better than it is:

- **A translation you have reviewed, accepted and published** is work delivered. The seven-day window
  still applies to it; beyond that, it is not refundable simply because you changed your mind later.
- **Machine translation makes mistakes,** and that is what the review tools are for. A specific
  translation being wrong is a good reason to ask for a refund inside the window - it is not a
  standing claim afterwards.
- **The free demo costs nothing,** so there is nothing to refund.

## Your statutory rights

Nothing in this policy takes away rights you have by law. If your local consumer law gives you a
longer or stronger right than the seven days above, that law wins.

## Contact

[aixlifftranslate@gmail.com](mailto:aixlifftranslate@gmail.com)