Aligner and More: Memsource Cloud 3.6 Released

Memsource Cloud 3.6 has been released on Sunday, 18 August 2013 at 8:30 AM GMT. Memsource Cloud 3.6 includes over 100 improvements and bug fixes. These are the most important new features:


  • With this release we are introducing a much-requested feature: an alignment functionality that makes it possible to create a translation memory from a translation (and its original) that was not processed through a CAT tool
  • To use this feature is simple:

Project Management

  • Memsource Cloud UI is also in Spanish now
  • New user detail page
    • We have introduced a new user detail page that, among other things, lists 30 most recent jobs assigned to the user
  • Delete all translations
    • The feature “Delete All Translations” will now delete all translation except locked segments
  • Filter by status in preceding workflow step
    • It is now possible to filter jobs by whether they are completed or not in the preceding workflow step

Translation Memory

  • Automated tag renumbering for higher leverage
    • For 99% TM matches because of different tag numbers from the segment for translation, we will automatically renumber the tags on the fly, so that the user gets a 100% match (in an analysis, TM search, pre-translation - simply everywhere).
    • For example, if your segment for translation is {1}Good {2}morning and the TM match is {2}Good {3}morning, you will get a 100% match now and will in fact see already the renumbered segment in TM results.
  • Export TM as XLSX, edit it and re-import it back
    • Translation memories can now be exported as XLSX, edited in MS Excel or similar application and re-imported back into MemSource
    • Memsource will update records that have been changed
  • Export only selected target languages
    • It is now possible to select which target languages should be exported for multilingual TMs

File Conversion

  • Improvement of HTML subfilter for XML files
    • When XML includes some HTML content, the HTML content can be processed by a HTML subfilter
    • Now it is possible to use the HTML file filter settings when creating a new job to customize the HTML subfilter
  • XLIFF segmentation optional
    • Previously, XLIFF files got segmented by default. Now there is an option that makes this optional
  • Non-translatable inline elements in XML
    • We have split the previous XML file import option  ”Inline elements” into “Translatable inline elements” and “Non-translatable inline elements”

Machine Translation

  • NICT MT engine


  • Download analysis as CSV/LOG file
    • A new API call analyse/download makes it possible to download the analysis files
    • However, if you need to get the analysis in a machine-readable format, the JSON response may still be your best option, via analyse/get
  • Get analysis
    • A new API call analyse/get makes it possible to get analysis results from an existing analysis (without the need to launch a new analysis)
  • Edit project
    • We have introduced a new API call project/edit that makes editing project data possible, such as updating project name, client, domain, adding target languages
  • List of jobs assigned to a user
  • Get progress
    • A new API call getSegmentsCount  makes it possible to get job metadata related to job progress and other job-related information, such as:
      • Number of segments
      • Quality assurance status
      • Translation progress
  • API token expiration
    • In line with our  API documentation we have started expiring API tokens
    • Each API token is valid 24 hours only
  • Client, Domain, SubDomain APIs
  • Get user by username
    • This new API call (user/getByUserName) makes it possible to get a user by his/her username