Hi how are things.
I have an EN-FR memory that contains some translation units in English (target language), which is incorrect. I need to remove those translation units from memory. How can I apply a filter to the memory so that only those units appear and delete them, without having to search the entire memory?
thank you so much.
Hi Alma Romo
Using the "Translation Memories" view in Trados Studio, you can either perform a search based on the content of the target segment or create a condition to perform a filtered search.
Here are the steps to perform a filtered search. If you provide more information about what you want to filter by, I could give a more specific answer.
Do I understand correctly that you want to automatically check which TUs have English language in the target field? Unless you manage to hook your TM up with some kind of AI process, there are simple ways that require some manual work:
Filter segments like Jesse Good explained. If a target segment contains any letter that is unique to French, filter it - no need to look at it. You might use a regex like this: À|à|Â|â|Æ|æ|Ç|ç|É|é|È|è|Ê|ê|Ë|ë|Î|î|Ï|ï|Ô|ô|Œ|œ|Ù|ù|Û|û|Ü|ü|Ÿ|ÿ
Depending how large your TM is and how many segments pass through this first filter, you could filter for the positive existence (regex) of frequent English words like: \bthe\b|\bbe\b|\bto\b|\bof\b|\band\b|\ba\b|\bin\b|\bthat\b|\bhave\b|\bI\b|\bit\b|\bfor\b|\bnot\b|\bon\b|\bwith\b|\bhe\b|\bas\b|\byou\b|\bdo\b|\bat\b
As well as frequent English letter combinations: "sh", "ck", "wh", which might not be frequent in French.
It would look a bit like this:
My French is nothing to write home about, but since you will know both languages, you will know what is specific to one of them.
Translate
