Fix a TMX that won't import

Q: why does Trados say “invalid culture identifier”?

usually because the TMX header has srclang="*all*", which the TMX standard allows but Trados doesn't read. turn on “fix language codes” and the header gets the language every unit starts with.

Q: will you change my translations?

only to make the file valid, and the preview lists every change: illegal control characters removed (Word's line break becomes a space), a stray & or < escaped, HTML entities like turned into the character they stand for. the words stay as they are.

Q: what if a unit is too damaged to repair?

it's left out of the cleaned file and quoted in the preview, raw, so you can fix it in your CAT tool. everything around it imports normally.

A TMX that won't import is rarely about the translations. It's one invisible character, a header that says srclang="*all*", the same language spelled three ways, or a closing tag without its opening tag. TM Cleaner finds those, repairs what can be repaired, and tells you exactly what it changed.

do it now →see a live demo

━━ the problem

one bad character stops the whole file

A vertical tab pasted from Word, a stray "&", an   or a byte in the wrong encoding makes the file invalid XML, and strict readers give up on all of it. Re-exporting gives you the same file back.

the language codes don't match

Trados can't open a TMX whose header says srclang="*all*". Phrase and memoQ match locales strictly, so units marked en_US, EN-US or plain en next to en-US can be skipped or land in the wrong TM.

units vanish without an error

Trados skips units where a closing formatting tag has no opening tag ("TagAnchorNotOpen"), and the import log just shows a smaller number. You find out weeks later, from a missing match.

━━ how TM Cleaner handles it

Drop the TMX in. Encoding, illegal characters, stray & and <, HTML entities and a cut-off ending are repaired before the scan; units too damaged to read are set aside and quoted so you can fix them. The “Will it import?” check then lists language-code and tag problems with one-click fixes, written into the cleaned file.

✓repairs the file instead of rejecting it: every change is listed, nothing is guessed
✓standard language codes, a real header srclang and TMX 1.4 xml:lang on export
✓units with broken tag pairs removed into removed.tmx, where you can fix them
✓“Convert only” keeps every unit: repair and fix codes without cleaning anything

━━ frequently asked

why does Trados say “invalid culture identifier”?+

usually because the TMX header has srclang="*all*", which the TMX standard allows but Trados doesn't read. turn on “fix language codes” and the header gets the language every unit starts with.

will you change my translations?+

only to make the file valid, and the preview lists every change: illegal control characters removed (Word's line break becomes a space), a stray & or < escaped, HTML entities like   turned into the character they stand for. the words stay as they are.

what if a unit is too damaged to repair?+

it's left out of the cleaned file and quoted in the preview, raw, so you can fix it in your CAT tool. everything around it imports normally.

━━ related pages

Translation memory cleaner for Trados Translation memory cleaner for memoQ Remove duplicate translation units from a TM Translation memory cleaner for Phrase (Memsource)Clean a TMX for MT training data Clean a translation memory for AI translation Translation memory cleaner for XTM Cloud Remove machine translation from a translation memory Translation memory cleaner for OmegaT Translation memory cleaner for Smartcat Translation memory cleaner for Crowdin Merge TMX files into one translation memory Convert Excel or CSV to TMX Reverse a TMX or extract one language pair Remove a bad import from a translation memory Anonymize a translation memory before sharing it guide: how to clean a translation memory

clean your tmx now →

free · no account up to 50 MB · files up to 2 GB · try the demo