harveyai/Recognizers-Text ? reverse-engineered prompt

Build me a text recognition library that can pull things like numbers, percentages, units, and date and time from plain text, and return a clean normalized result. It should work well in multiple languages, especially English, Chinese, French, Spanish, Portuguese, German, Italian, Turkish, Hindi, and Dutch, with basic support for a few others if that is already in the project.

Please wire it up so it is easy to use from .NET, and keep the design friendly for the other language versions already in the repo. Add solid tests using the existing spec style, cover common examples and edge cases, and make sure the README has a simple usage example so someone can try it right away.

If you need to check current docs or examples online to match the right behavior, go ahead and look them up.