Re: New Word Generator
From: | Petr Mejzlík <imploder@...> |
Date: | Sunday, September 30, 2007, 17:58 |
Dne Sat, 29 Sep 2007 23:30:26 +0200 taliesin the storyteller
<taliesin-conlang@...> napsal/-a:
> * Petr Mejzlík said on 2007-09-29 22:10:31 +0200
>> One that could learn automatically would be amazing, but I
>> guess that's too far-fetched and really complicated to make.
>> OTOH people can get used to a new phonotactics intuitively so
>> a computer should be able to do it as well (and it's not about
>> meaning so it shouldn't require human knowledge). It would
>> have to somehow identify and classify the phonemes with some
>> really neat statistical algorithm. Not that I know how to do
>> it though.
>
> What you want is a Markov generator. Look up "Markov chain" on
> Wikipedia. They are commonly used to make nonsense text and spam
> text, though it's a no-brainer to have it learn phonemes instead
> of words. Give a Markov-generator some words and it'll output
> words that resemble the input. The problem though is how the
> quality and quantity of input affects the quality of output; for
> conlang word-generation it is important to have enough input, if
> necessary by duplicating some words.
>
>
> t.
>
It might suffice to persuade the author of this:
http://www.fourteenminutes.com/fun/words/index.cgi
to make it accept any text to base the words off you give it. The English
words it makes now look good.