Translate XML files. Custom schemas. Tags untouched.
Drop your .xml — DITA, DocBook, custom CMS export, configuration file, or proprietary schema. Fily detects translatable text, leaves everything else alone, and respects your schema.
What is XML (Generic)?
XML is the universal structured-data format. Translation tools need to know which text inside an XML file is translatable (paragraph content, headings, captions) and which is not (IDs, references, code, configuration values). Different XML schemas — DITA, DocBook, custom CMS exports — have different rules.
Why XML (Generic) is tricky for AI translation
- Schema detection: a generic AI translator either translates everything (breaking IDs and references) or nothing (missing translatable content).
- translate attributes: many XML schemas use translate="no" or its-translate attributes to mark non-translatable subtrees. Naive tools ignore these.
- CDATA sections: text inside <![CDATA[]]> may need translation or may be code — needs schema-aware handling.
- Element-level instructions: DITA's <ph keyref="..."/> references are not translatable; DocBook's <code> blocks are not. Schema knowledge matters.
- Custom namespaces: client CMSes invent their own namespaces with their own translate flags. Generic tools strip them.
How Fily handles XML (Generic)
- Schema auto-detection: DITA, DocBook, TBX, TMX, OPC variants recognized automatically.
- translate="no" and its:translate="no" honored: subtrees marked non-translatable are passed through.
- Element-class translation rules: per-schema rules for which elements carry translatable content vs structural data.
- Custom namespace pass-through: unknown namespaces and attributes preserved unchanged.
- CDATA detection: text vs code classification with conservative defaults; user override available.
Pipeline: xml_qa_12step_v2@1.0.0
The XML (Generic) workflow with Fily
Upload
Drop your .xml (single or batch ZIP). Optional: glossary, TM, style guide.
Process
Fily runs the XML (Generic) pipeline + 12 QA steps. Typical job: 10–20 minutes.
Download
Same format, ready to deliver. QA report HTML attached.
Common upload: a documentation export from a CMS in DITA, a DocBook manual, or a custom XML from a client's internal system. Fily delivers the same XML with translatable text translated, structural elements untouched, schema validation passing.
Beyond the standard pipeline
What we've built around XML (Generic)
Edge cases clients brought us for this format — and the pipelines we shipped to solve them.
4-layer cascade
Multi-tenant glossary cascade
An LSP with 40+ end-clients needed org-level terms, client-level overrides, and sub-client (brand) overrides to all stack — without re-attaching glossaries to every job.
A 4-layer cascade: system → org → client → sub-client → job. Forbidden terms inherit and override, style guides cascade, and every future job picks up the right rules automatically.
Golden TM · auto-build
A translation memory that builds itself
Clients wanted TM leverage but didn't want to maintain a TM — exporting, cleaning and re-importing TMX after every project is a job nobody has time for.
Every confirmed segment is promoted into the organization's Golden TM automatically, scoped per client and language pair, and reused on every future job. Reviewer edits in the dashboard feed straight back in. Exact matches then cost $0.
Frequently asked about XML (Generic)
Other formats Fily translates
Ready to translate a XML (Generic) file?
No card. No setup. Upload one file and see the output.