This is not the first time I have seen these so-called "fuzzy" matches (see example below). I am currently running Studio 2017 SR1.
Does anyone know why Studio thinks the words below have changed?
This is the segment as displayed in the Translation Results window. Studio sees the rest of the segment as unchanged.
Or am I the only one seeing such things?
Best regards,
Bruce Campbell
ASAP Language Services
I found the answer. The document I am translating was converted from PDF and has two forms of vowels with umlauts, the regular umlaut vowels from the font itself and others that appear to be constructed from the vowel and two dots.
I am not sure how the strange constructions are formed, but here is what I saw when I copied one of the strange "ü" characters in the Word file and pasted it into the Word search box.
Note how the dots over the "u" are shifted to the right in the Word search box, which makes me think these strange "ü" characters have been somehow constructed from a "u" and two dots.
Studio recognizes that the two are different, so the difference between the words in the segment above is that either the "ü" in the segment or in the TM is one of these constructed characters.
I will have to ask the client how the PDF conversion was done and how to avoid this in the future. Otherwise it is impossible to reliably search for words containing umlaut characters.
Best regards,
Bruce Campbell
ASAP Language Services
Translate
