VOP currently has the following language processing capabilities:
Built-in support for languages
- base language packEnglish (eng), Japanese (jpn), Korean (kor)
- hybrid recognition: By
--lang eng+jpnParameters for simultaneous processing of bilingual documents
Expanding New Language Steps
- Install the corresponding Tesseract language packs (e.g. Chinese:
sudo apt install tesseract-ocr-chi-sim) - compiler
config/languages.jsonAdd a configuration item:"chi": {
"tesseract_code": "chi_sim",
"google_vision_code": "zh"
} - Restart the program and specify the language parameter:
--lang chi+eng
Caveats:
- New languages need to be tested for recognition accuracy, and complex layout documents may require adjustment of YOLO model parameters.
- Non-Latin languages (e.g. Arabic) require additional configuration for text direction detection.
- It is recommended to prioritize the use of Google Vision API (200+ languages supported), but its recognition accuracy for formulas/tables may be reduced
This answer comes from the articleVOP: OCR Tool for Extracting Complex Diagrams and Math FormulasThe
May not be reproduced without permission:AI productivity tools " What languages does VOP support for text recognition? How to extend the support for new languages?
English 
简体中文 
日本語 
Deutsch 
Português do Brasil