Closed nullren closed 11 years ago
That's the issue with ISO 639-2/3. chi and zho are the ISO 639-2 identifiers. I don't know why tesseract doesn't use ISO 639-1 (zh) instead which is simple, commonly used.
Anyway I added an option --tesseract-lang
which you can use to specify the tesseract language to use.
8d34a407b326
in my subtitles file, when i listlangs, i get "zh", so when i select "zh" with
--lang
, vobsub2srt looks forzho.traineddata
in tesseract. it should be looking forchi_*
or perhaps i should be able to specify this thing directly.