search

SubtitleEdit / subtitleedit

the subtitle editor :)
http://www.nikse.dk/SubtitleEdit/Help
GNU General Public License v3.0
8.86k stars 916 forks source link

OCR .sup Tesseract + LSTM detect Italic #9012

Closed Slava46 closed 4 days ago

Slava46 commented 6 days ago

Is it possible to do italic detection when do OCR .sup using Tesseract + LSTM? Because just Tesseract 5.3.3 can detect italic, but result of OCR'ing much better with using + LSTM mode. And for more quality mode Tesseract + LSTM not working italic detect...

niksedk commented 4 days ago

Sorry, Tesseract LSTM is just bad at detecting font properties :(

You can try nOCR, which works well for subtitles with large fonts.