A couple of minor suggestions

[Ivanx32]

1. Perhaps googletrans library can simpify the process. You do not have to translate each part the original document and then combine translations.

2. I think in some cases automatic fixing of translated tokens is possible:

• [2.77 ] // extra space; very often case
• {2.353] // different barackets
• [2,223] // comma instead of dot; as far as I understand, already supported
• [2238] // missing dot; (I'm not absolutely sure on this case)

3. Paste the code into .tex file:
   \usepackage[utf8]{inputenc}
\usepackage[russian]{babel}
   for Russian (or other) language support or mention about this in Readme.