A few weeks ago, I added all NEU tags of “Bleistift” to removed.txt because only “der Bleistift” is correct. The next day, when I entered “Bleistift” into the text analyzer, it was only tagged with the MAS tags. But now, the analyzer tags Bleistift as “MAS” and “NEU” again. What happened?
Are you absolutely sure it ever worked? The thing is that “Bleistift” is not known as a word, thus it’s analyzed as “Blei” + “Stift” (and “Stift” has a NEU
reading). But as far as I can see, this has been the case for a long time. Anyway, it might help to add “Bleistift” to added.txt
so no compound analysis is needed.
Quiet sure but not absolutely sure. I’ll try added.txt, thanks for the hint!
Suggestions don’t work as expected. It still suggests “das Bleistift”.
Maybe it should be generally prevented that LT suggests the same string that it has marked as incorrect?
1 Like
That’s probably because the synthesizer uses its own dictionary and doesn’t consider removed.txt
.