Skip to content

Latest commit

 

History

History
61 lines (38 loc) · 1.75 KB

README.md

File metadata and controls

61 lines (38 loc) · 1.75 KB
Raw

SnapStudySensei

SnapStudySensei is a tool to assist with capturing, extracting, translating and recording Japanese flashcards into Anki.

This project is inspired by Game2Text but takes a different technical approach.

SnapStudySensei screenshot

Known limitations

  • Currently only supported on Linux. Porting it to macOS and Windows should be doable by adding the ability to list windows (see windows_list.py, patch welcome)
  • No configuration, only designed for my own needs so far
  • A bit slow to start due to the OCR model initialization

Installation

Anki and its Anki-Connect plugin must be installed.

python -m venv venv
. venv/bin/activate
pip install -e .

Important note

This is a non-intrusive standalone installation, which means the Qt libraries are duplicated within the virtual env. This breaks at least the fcitx input method setup on the system. Setting IBUS_USE_PORTAL=1 QT_IM_MODULE=ibus in the environment can be used as a workaround (ibus doesn't need to be installed).

Running

Anki and its Anki-Connect plugin must be running.

Enter the venv (. venv/bin/activate) if you used that installation method, and run sss.

Important note

SnapStudySensei will automatically create a model, deck and flashcard templates in Anki. The deck is called SnapStudySensei and is located in the Japanese category.

Thanks to