Ask HN: What have you used natural language identification tools for?

I am curious about real world use cases for natural language identification. If you have used language ID tools before, what was your use case? I would like to hearing about: - How much text/data you were dealing with - What tools or libraries you used - Whether the result was good enough in production or only for preprocessing - If the performance, speed, of the tool was a problem - Any common problems you ran into