Corpus-analyzer is a project mainly written in Python, it's free.
A webapp that takes a piece of your writing, generates word frequencies, and compares them to the British and American National Corpora.
My website is now offline. I've migrated the front page to github pages and the blog to tumblr. But there is no where for the Corpus Analyser to go.
Therefore, I am putting a standalone version of it here on github. The only dependency is web.py.
Type python start.py or just ./start.py to run the application.