26th January 2022

Expertise as localizer and post-editor

I started my translation business in 1997, so this is my 25th anniversary. My language combination is English into Spanish, I haven’t used my language C (French) for ages, and my areas of expertise are Life Sciences, Marketing, IT and Technology, but mainly Life Sciences. My first contact with MT was many years ago (10-12?), when Lucía asked me to do a test to compare a couple of engines vs. human translation. After that, I have passed several training courses with different clients in order to meet their expectations (light PE, full PE), but mainly you learn with practice.

How do you mostly use MT? Do you use it for particular language combinations and content types, and what aspects affect your decision on whether to use MT or not?

I use MT for EN>ES and mainly to translate Clinical Protocols and medical devices IFU. I normally use MT in 2 ways:

• MT engine is integrated into the CAT tool, and I accept and amend MT suggestions as needed (one of my clients has their own CAT tool with 4 different engines integrated, so I can choose if any of them is suitable and insert the translation or translate from scratch)
• File is already pre-translated with MT and TM when applicable, and I post-edit the result

PE is not suitable for marketing, for instance. I normally check that the documents don’t include many tags, software strings, very specialized terminology, format issues like tables (to avoid segmentation issues), etc.

Post-editing requires a lot of corrections that are repetitive, mainly deletions, insertions and substitutions (punctuation, gender/grammar concordance, tags…). All these repetitive edits can cause strain in our body and can also be psychologically draining. What tips and tricks do you have to avoid such physical and psychological issues?

First of all, I have to say that my physical environment is quite ergonomic, with several screens and a vertical mouse, among other things. For the PE specific tasks, I mainly use filters to make sure that the number of products are not translated, in Medical documents there are lots of acronyms which the engine usually translates incorrectly or the same acronym can appear translated in several ways, etc. Of course I use many shortcuts, search and replace (using the filtering first, to make sure I am change just wat I want to change), regex (to look for numbers, for instance), QA tools such as Xbench, AHK… Now I am getting used to Mac shortcuts too!

Research shows that CAT tools positively contribute to a more efficient post-editing but based on the comments from some of the translators questioned there is still room for improvement. How to get away from segmentation, for example, is one of the most commented aspects. What type of shortcomings do you see in CAT tools?

It would be great if we could block specifically product names and those terms which should remain in English, or to find a way to automatically insert the translation of software strings (in tags?). Also, there are still some CAT tools which don’t allow to open several files at the same time, which improves consistency… or that all these CAT tools could be used in Mac!

Back to top Previous page

Back to Table of Contents Next page