Jun 22, 2017

Translation alignment: the memoQ advantage

The basic features for aligning translated texts in memoQ are straightforward and can be learned easily from Kilgray documentation such as the guides or memoQ Help. However, there are three aspects of alignment in memoQ which I think are worth particular attention and which distinguish it in important ways from alignments performed with other translation environment tools or aligners.

Aligning the content of two folders with source and target documents; automatic pairing by name
The first is memoQ’s ability to determine alignment pairs automatically based on the similarity of names. This has the advantage, for example, that large numbers of files can be aligned automatically, with the source and target documents matched based upon the filenames. This can be done with individual files or with entire folders with perhaps hundreds of files. Thus if source files are contained in one folder and the translated files in the target language are in a different folder and the source and target file names are similar, the alignment process for a great number of files can be set up and run in a matter of minutes. Note in the example screenshot above that different file types may be aligned with each other.

The second important difference with alignment in memoQ is that it is really not necessary to feed the aligned content to a translation memory. memoQ LiveDocs alignments essentially function as a translation memory in the LiveDocs corpus, with one important difference: by right-clicking matches in the translation results pane or a concordance hit list, the aligned document can be opened directly and the full context of the content match can be read. A match or concordance hit found in a traditional translation memory is an isolated segment, divorced from its original context, which can be critical to understanding that translated segment. LiveDocs overcomes this problem.

A third advantage of alignment in memoQ is that, unlike environments in which aligned content can only be used after it is fed to a translation memory, a great deal of time can be saved by not “improving” the alignment unless its content has been determined to be relevant to a new source text for translation. If an analysis shows that there are significant matches to be found in a crude/bulk alignment, the specific relevant alignments can be determined and the contents of these finalized while leaving irrelevant aligned documents in an unimproved state. Should these unimproved alignments in fact contain relevant vocabulary for concordance searches, and if a concordance hit from them appears to be misaligned, opening the document via the context menu usually reveals the desired target text in a nearby segment.

Concordance lookup in memoQ with direct access to an aligned document in a LiveDocs corpus


  1. I still wonder why PDF files are not imported by default as plain text. It requires 4 additional clicks every time you want to import two PDF files. Import as docx with hard returns is useless for alignment purposes. I made a suggestion to change this but nothing ever happened.


    1. Noe, all of the PDF import filters for memoQ are total crap. If you want to extract and align PDF content I suggest using the XML export from Iceni InFix.

    2. Hi Kevin, thanks for your reply. Don't know if I am willig to pay, even it would be just some cents, for every alignment passing through Iceni.

    3. What are you talking about Noe? pay for what? Using the XML extract function of iceni InFix is free, just like the TransPDF service they have for extracting XLIFFs for translation. In fact, AFAIK, the only function in InFix which is actually crippled if you don't have a license is the useful "story export" feature, which is rather nice for pulling PDF content out as a relatively clean MS Word file. To do alignments or stick content into LiveDocs for reference you are not subject to any fees for removing watermarks from PDF files created with InFix (which is actually not relevant). Another nice thing about the iceni tool is that it ignores password protection on a PDF file, which can also be useful.

    4. Hi Kevin, I just know upgraded to 8.1.4 so I'll give the Iceni XLIFF extraction a try. I thought it's a payable service. ;)

    5. The XLIFF is for tranalation, not alignment. And in some cases people may run afoul of their NDAs using that online service with the latest version of memoQ. But for alignment and confidentiality situations, the older form of Iceni translatable export can be used: the XML files. That is done directly in InFix. I think I wrote an article about this some years ago, but the menus have changed since then, and I intend to write a short post soon with the new details. It's a bit hard to find the relevant menu commands, especially if one is averse to RTFM.


Notice to spammers: your locations are being traced and fed to the recreational target list for my new line of chemical weapon drones :-)