Apply New Translation Memory and Terms Against Old Translation Document

Hey there!
I'm new to this software, and I'm in the process of a trial.

I have a very specific use case, and I'm hoping to find an answer to it.

Recently, a small team and I have been working to translate Spanish texts into English.
I have now created a small translation memory from these checked translations.

Since we are such a small team, and we are working completely voluntarily for no revenue, but we still want consistency, we figured this tool would be great for revising old translations that are not so great.

My hope is that I can take this Translation Memory bank and apply it against old English translations in order to automatically fill in specific glossary/term-base items, as well as common phrases that have been corrected in our newly-verified translations.

There are many terms and phrases in the Spanish that repeat, and have a very specific meaning, and it would seem as though we could run some task on the previously translated English in order to 'correct' it.

Am I correct in thinking that SDL Trados Studio can do this task for me without having to re-translate the entire document?
Am I able to load in a source document, and a previous translation, and simply work through it as such as a starting point (fully aligned)?

If you could point me in the right direction to articles, videos, etc, and assure me that this is indeed possible, I truly would be extremely grateful for your help, as it will assist this small non-profit team of volunteers achieve a great mission.

Thank you so much in advance!!

  • no. it sounds almost impossible to me
    you have to have 1:1 translated terms list
    SDL Trados Studio is the most expensive tool in its sector
    you could find lot of nice tools out there, "Free of Charge"

  • Hi  


    Perhaps you could align the previous English translation with the new ones you have corrected and create an English (old) to English (new) Translation Memory (in practice you'd have to use en(GB) to en(US) to get around the fact you can't go from en(GB) to en(GB) for example).  You could then use your new Translation Memory as leverage to help you with your old English translation projects.  What will happen is this:

    • you might get some automatic 100% matches that will automatically populate as a result of you having translated the same sentence into your new Translation memory.
    • you might get high fuzzy matches which "might" automatically repair themselves if you have enough data in your Translation Memory, and perhaps have a glossary to use too.
    • you might get low fuzzy matches that you'll have to repair yourselves
    • you might get no matches at all in which case you would have to translate them yourself
    • all the time your Translation Memory will hopefully be getting more and more useful as time goes by

    But that's all theory and the principle of how these tools work.

    You could also build a glossary English (old) to English (new) and use this help you repair the old English and bring some consistency into it's use.  Again it won't do this for you automatically but you could search & replace, run terminology verification to find missing incorrect terms etc.

    If I was you I'd have a play with the trial and see how this might be able to help you and make your minds up after that.  If you need some help to try this in practice I'd be happy to take a call with you and at least show you how this might work in theory?