Just brilliant.

What about OCR in realtime video mode? Is it possible?

@pulbrich Now it works in both directions :-)

@cvp I was qurious about the script, but I received "The file pyto.py couldn't be found", when Pytonista opens.

Vision OCR produce prety good results.

But has someone succeeded to OCR text with indexes or power signs?

(https://commons.wikimedia.org/wiki/File:Quadratic-formula.jpg)

@JonB Good note, thank you.

I tried it.
In my first cell I had to download a file, using wget. Unfortunately it was not recognised as valid command.

BTW, I'm impressed from accuracy ( for Latin based texts ).

Doesn’t work for Cyrillic (Bulgarian).

I think VNRecognizeText works only for English.