I'd just heard about mathpix, a way that can identify formula from images and generate the LaTeX code. I have some handouts (already printed) from my teacher 10 years ago written in Chinese and many math formulas. I don't have the original digital file, but only have those documents on my shelves. I want to turn it into digital files, namely get the Chinese texts and math formula in LaTeX code so that I can reproduce and reprint it. However doing it by hand is a heavy work, so I want to seek some clever way. I think mathpix can help me a lot. But I have two main questions with respect to it:
If I have a picture like this, with many inline math: (just a demo, not the actual document I have)
Can I get the result both with the English words and LaTeX inline math? (I mean, get the resulting string "
Suppose $A$ is bounded subset of $\Bbb R^n$. If ...
") It seems that I need a pure text OCR tool and mathpix work together nicely. How to achieve such task?If I have bunches of images to identify, I guess I need to write some python program with the mathpix API provided in mathpix API. But the sample code given is not work in my python 3 now. I'm not good at python, how to modify it? Or is there other clever way to do? (Maybe I should ask this question in another board, but I think it would be fewer people know LaTeX there.)