Given some sample text, say Shakespeare, as input, the computer generates output which is random, but which has the same statistical distribution of characters or combinations of characters. (A character may be a letter, a digit, a space, a punctuation mark, etc.)
You can tell it how many characters to match on, from one to six. You need to pick at least level 3 to get anything resembling words, but the accuracy of word choice jumps dramatically from 2 to 3, and then again from 3 to 4. I did all of my experiments on level 6.
I started by inserting the complete text of this post about a concert week, and it gave me this:
Of course, it was Carmina Burana for the second verse, it was a good feeling concert itself went on and learn. The sing one of the back of having a verse in a venue that are distracting! I kept have a fairly grueling concert weeks go. First rule of stores, and we're getting up several years. But it's part, and do it all that none of the heck ever experience. The conductor cued then set up in a couple of stores, and voice.) So they make you ask to stand (stand and that these think it's not the new maes
Then I tried some fic, Secret of the Flames, and came up with several gems, including:
I’d like ages ago. Lilli’s hand, shook his corner and night, then whipped into a small grin. “Not enough to risk gettin’ caught. C’mon Lilli’s hand, she fear running monk nodded. “Shut up, then slipped out of its other shout broken only by torches and the fayth.” Paine could say a word. “I dunno, Paine’s guard had assume that she stone door, they’re still in the aeons were you promised…” “That were you thinking?” He half-dragged her balance fizzled into a small gripping her arms. “I’d like to see him,
Fun stuff. Check out the samples.