- Recently I use this repo to extract the subtitle, but I found the Tesseract is very hardly to use. - I combined RapidOCR and the code of this repo, made a simple transformation, and made it easier to use. - The specific link is as follows: https://github.com/SWHL/RapidVideOCR
https://github.com/SWHL/RapidVideOCR