I recently used CT to translate a fairly short document in Word (2700 words including 12 footnotes). I selected the option to segment the document into single sentences - because the sentences were long and convoluted.
My Java is set to the maximum permissible for CT.
CT decided to break down not only all the footnote numbers as segments, but also every pair of inverted commas (quote marks) as segments.
By the time I got about halfway through the translation things had got so slow that I abandoned CT and finished the job manually.
What was I doing wrong?