Placing tags using Auto-Insert Tags - SR1 Trados Studio 2021

Good afternoon,

I've just installed the SR1. I wanted to try "Placing tags using Auto-Insert Tags" but it doesn't work. 

I followed tutorial, enabiling Repair, checking if my TM has more than 1000 units, but anyhow it doesn't work. 

Where could be a problem?

Thanks! 

Michaela

Parents
  • I think this feature is indeed going to raise many questions like this.  It's a tricky topic so we've tried to explain as well as we can why you'll find that sometimes it works and sometimes it doesn't.  There isn't an easy way to show you why your specific example may not work.

    Like fragment matching and MatchRepair, automatic tag placement is built on upLIFT alignment.  Just like those earlier features, a TM with 1000 TUs is the minimum at which you can expect any alignment-based functionality to start working.  However, a TM of that size provides relatively little information about word relationships, so relatively few words can be aligned with confidence.  The aligner used for upLIFT attempts to use only high-confidence alignments, since lower-confidence alignments are more likely to be incorrect, and results based on those alignments (poor fragment translations, repairs or tag placements) will just get in the translator’s way.  So we take the view that it's better to produce nothing than to produce results that waste translator time.  A larger TM generally provides more information about word relationships, so produces more high-confidence alignment coverage.  Just the same, the current aligner primarily favours speed (so as to be suitable for desktop use), so may still lack coverage or accuracy in larger TMs in some cases.  It’s likely that an improved aligner will be available in a later update.

    So all in all this is the first release of a developing technology and we are actively working on improving this to try and provide a consistent, accurate result when attempting to use this for automatic tag placement.  In the meantime keep working with it, and as your TM grows and you are working with material where we can create aligned phrases that help with the text you're working on, you should start to see the benefit.

    Paul Filkin | RWS Group

    ________________________
    Design your own training!

    You've done the courses and still need to go a little further, or still not clear? 
    Tell us what you need in our Community Solutions Hub

Reply
  • I think this feature is indeed going to raise many questions like this.  It's a tricky topic so we've tried to explain as well as we can why you'll find that sometimes it works and sometimes it doesn't.  There isn't an easy way to show you why your specific example may not work.

    Like fragment matching and MatchRepair, automatic tag placement is built on upLIFT alignment.  Just like those earlier features, a TM with 1000 TUs is the minimum at which you can expect any alignment-based functionality to start working.  However, a TM of that size provides relatively little information about word relationships, so relatively few words can be aligned with confidence.  The aligner used for upLIFT attempts to use only high-confidence alignments, since lower-confidence alignments are more likely to be incorrect, and results based on those alignments (poor fragment translations, repairs or tag placements) will just get in the translator’s way.  So we take the view that it's better to produce nothing than to produce results that waste translator time.  A larger TM generally provides more information about word relationships, so produces more high-confidence alignment coverage.  Just the same, the current aligner primarily favours speed (so as to be suitable for desktop use), so may still lack coverage or accuracy in larger TMs in some cases.  It’s likely that an improved aligner will be available in a later update.

    So all in all this is the first release of a developing technology and we are actively working on improving this to try and provide a consistent, accurate result when attempting to use this for automatic tag placement.  In the meantime keep working with it, and as your TM grows and you are working with material where we can create aligned phrases that help with the text you're working on, you should start to see the benefit.

    Paul Filkin | RWS Group

    ________________________
    Design your own training!

    You've done the courses and still need to go a little further, or still not clear? 
    Tell us what you need in our Community Solutions Hub

Children