There are currently 4 Python NLTK demos available. On the top left you can do sentiment analysis, which uses text classification to determine sentiment polarity. On the top right, you can see how different word tokenizers work. On the bottom left, you can try stemming text in 17 supported languages. On the bottom right, you can do part of speech tagging with 22 different part of speech taggers, along with the possibility of chunk extraction and named entity recognition. Enter some text and see what happens!
If you'd like to use any of this functionality thru an API, checkout the Text Processing API Docs. And for higher limits and premium API access, signup for the Mashape Text-Processing API.
If you answered yes to any of these questions, please fill out this Natural Language Processing Services Survey.