markovify

there are several nice libraries that allow you to generate plausible sounding gibberish this one is particularly nice and easy to use: https://github.com/jsvine/markovify/ you give it a file of existing text and it generates complete rubbish that would pass most automatic filters

Not exactly concerning image AI, but regarding LLMs -- in case your site does not depend on indexing -- you can easily generate boatloads of garbage text via Markov chains and invisible-link them. One cool project for that> https://github.com/jsvine/markovify (Personally had the most fun outputs when constructing the corpus from technical articles mixed with esoteric babble) Active defense!

Finally, a concise explanation why this generation of chatbots –while absolutely superior to previous art (where applicable)— still throws me back to discovering markov chains*, 'cat'-ing countless amount of text into a 'corpora.txt' to build models of various personalities. The idea that a larger corpora == better was quickly established, but I hit a ceiling quite soon on early RPi's. It was possible to seed a start word for text generation, a Baby prompt we might call it? Nice feature was being able to make a mini chain (model) of the user input sentence, then merge that statistics into the bots chain. The verbatim record of what had been said was obscured but some words and their relations —the experience— was kept. So I imagined, at least. * I recommend giving the 'markovify' README a view: https://github.com/jsvine/markovify#basic-us

An easy way to get started that doesn't use pure ML is Markovify [1]. I used it for a couple of projects last year. [1] - https://github.com/jsvine/markovify

What I would do is use the Markovify python library and feed it with several texts from Project Gutenberg... try to generate some Lovecraftian prose or something... https://github.com/jsvine/markovify

I have to fully agree. The language-processing part is really simple (partly due to the really cool markovify library [1]) You may also like my older project [2] even though it is partly in German. I'm using Markov Chains for the titles and some custom regex-based language processing for the descriptions. [1] https://github.com/jsvine/markovify/ [2] https://nonsense.lw1.at/

I bet whole source code of that engine can be found in Basic Usage section of https://github.com/jsvine/markovify

Yes! In 1948, Shannon proposed using a Markov chain to create a statistical model of the sequences of letters in a piece of English text and this model can be used to generate random text given some existing text. (http://www.cs.princeton.edu/courses/archive/spr05/cos126/ass...). Here is a GitHub implementation: https://github.com/jsvine/markovify Deep models like LSTM/RNN can probably produce better results.