work in progress: charNGram
I know there are plenty of character n-gram generators out there (Travesty, Dissociated Press, JanusNode, etc.) but there’s something enjoyable about coding one yourself. Among other things, I think it’s pretty clear that you can skip explicitly building a language model in memory; like Dissociated Press you can just use the text. At some point I should write a proof of the equivalence of the two approaches.
kn ich flomalveeane thin,
Whe pay s ld tharde whenl,
Looks like a mix of Sanskrit and German.