Dear All,
There are two lines in an email I might receive any day, lines that quickly draw my attention. The first is ‘Saluti da Sacro Cuore...’, because it is inevitably followed by a request for translation ????.
The other is a question: ‘How long would it take you to translate....?’
Let’s leave the first aside. That is merely an occupational hazard. But let’s take the other one and try to answer it if, say, we have an item to translate that is 5,000 words long, or say about 12-13 A4 pages. How best to work this out?
Consider some of the factors involved:
- number of words
- text specialisation
- whether or not you have the original file format (e.g. docx instead of pdf)
- your availability (since you may have many other things to do).
On average, a translator who is able to dedicate time to the task will be able to do 2,000-2,500 words a day. Assume something around 300 words an hour, but then you have to factor in review of what has been done or, if the text contains citations, checking each citation for its authenticity and accuracy. So the 5,000 word item will take at least 2 days, and if we are busy with other things, longer.
Now, all of the above ignores two other factors: the increasing sophistication and accuracy of NMT (neural machine translation) such as DeepL, if it works for your language pair, and the fact that you might be using a CAT program (Computer Assisted Translation), which also means you are using TM (Translation Memory). The use of either or both of these translation techniques can dramatically reduce the amount of time, and also the nature of the task, because especially in the case of NMT, you are doing post-editing for the most part. DeepL or some other similar engine does the bulk of the translation work and you make appropriate and necessary adjustments. And in the case of a TM, if what you are translating contains anything you have previously translated, and a glossary, then you will be prompted to make use of these (or translation units will be dropped in automatically if there is a sufficient match).
Perhaps next week I will take up the Translation Memory issue in more depth. If you are not using it, you should reflect on that and consider how it might be able to help you, and indeed, how we might be able to help each other. I have many I could share, and perhaps you have some too.
best_practice_27-Translation time.docx
