[Documentation] tessdata

[J Ano]

OCR is working amazingly for English subs.

For other languages, OCR is working OK, but not not as nice as English.

In the documentation (Wiki), it’s not specified which tessdata shall be used. For instance, should we stick with 3.02 files or can we use 4.0 files ?

I would be nice to have this documented. If you provide me with this info, I’ll write an article on the Wiki.

[Damiano Galassi]

3.02, I tried to update to 4.0 but it was 10 times slower. So for now I am sticking to the old version. There is already a link to the download page on the wiki https://bitbucket.org/galad87/subler/wiki/Subtitles%20Guide but maybe it can be made clearer.

[Damiano Galassi]

Actually I think it was slower because it defaulted to a different algorithm, I'll try it again soon.

[Damiano Galassi]

The next version will have a window to download the tessdata files.