Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Draco18s]

The program breaks words up into their component letters, keeping track of which letters each one is followed by.  For instance, "arrkhal" would yield

Markov Algorithm, ingenious.  (When you perform it on words in a sentence you get grammatically correct nonsense--I used it for an Iron Author contest once*).

*Input text: a 20+ page short story I wrote.  Output: 16 pages of gibberish (plus two pages copied out of the middle, pasted later on, font changed to ClaWrite^).  Number of judges forced into a drunken stupor: three.  Number of judges who held a grudge for a full year: one.  Length of ban**: two years.  Alternative punishment: serve as a judge (almost carried out).

^As readable as wingdings, if you've never seen it before.

**I was banned from a "write bad fiction" contest for writing fiction that was beyond bad: it was unreadable.  I plan to do better this year.  Buhahaha.

[RavingManiac]

The program breaks words up into their component letters, keeping track of which letters each one is followed by.  For instance, "arrkhal" would yield

Markov Algorithm, ingenious.  (When you perform it on words in a sentence you get grammatically correct nonsense--I used it for an Iron Author contest once*).

*Input text: a 20+ page short story I wrote.  Output: 16 pages of gibberish (plus two pages copied out of the middle, pasted later on, font changed to ClaWrite^).  Number of judges forced into a drunken stupor: three.  Number of judges who held a grudge for a full year: one.  Length of ban**: two years.  Alternative punishment: serve as a judge (almost carried out).

^As readable as wingdings, if you've never seen it before.

**I was banned from a "write bad fiction" contest for writing fiction that was beyond bad: it was unreadable.  I plan to do better this year.  Buhahaha.

Would you be so kind as to post some sample text? A lot of us want to know what grammatically correct gibberish looks like.

[Draco18s]

Would you be so kind as to post some sample text? A lot of us want to know what grammatically correct gibberish looks like.

I will dig some up when I get home.  For the most part you get Lorum Ipsum (which is neither English, gibberish, or made up).

[Samalander]

Anybody have a good place to get/create more language seeds from? This is awesome enough that I want to use it for my D&D game to generate the 'non-common' (IE elven, goblin, dwarf, etc.) languages.

Thanks!

Edit: I found my wife's Japanese-English and Latin-English dictionaries. I'd be willing to email the seed files to interested parties.

[Draco18s]

I will dig some up when I get home.  For the most part you get Lorum Ipsum (which is neither English, gibberish, or made up).

Quick note on this:
I've misplaced the output file I had already, so I'm going to need to make another, which is kind of tedius.

Blag.  What I thought was the compiled program was in fact blank and the source didn't recompile ("srand" wasn't declared; its supposed to be a standard library random seed function).

But I did find an online version of the same thing.  It output a few sentences at random, based on my input text (which was sufficiently large to produce gibberish).  Which is better for showing what the program can do than just letting it run until it hits an EOF character (as reading several pages of this kind of jibberish does tend to break your mind).

• He hurried down the first with his mental probe, its casing shattered like fine china falling on the march headed this way.
• This door was open or not, there wasn’t anything inside the maze through the endless maze coming to more splits, doors, empty rooms, and long hallways.
• I know almost nothing about the dragons are evil.” As Lothor spoke, he sent tendrils of his weight behind another charge, and the sight it shown upon was not supported by anything, and it broke free enough to hold probably ten dragons or so, he was at a jog, dodging between trees in order to make a good decision unless we know.” The Elder nodded once.
• Don’t be fooled though, we suspect that the hall towards the glowing spark.
• He continued down the dark, stone passageway claws clicking on the other things that would disrupt his concentration.
• He put his sneaking skills to use humanity’s power to destroy the “demon-spawn.” But Mother Earth prevented that; everything the scientists build failed to work.
• The two of them aside until he found himself in one of the cavern, now he had entered through and the cave.

[Igfig]

Version 1.7 has been uploaded.

This isn't the big one with the variable-order Markov generation model; that'll be a project for this summer.  What this one adds is the ability to extend existing translation files with new words, and to set bounds on the length of generated words.

Note that specifying a narrow range of word lengths can exacerbate the problems of having too few words in your language file.  Try to keep the same range of lengths found in the language file for best results.  (I'll put a thing that tells you exactly what those ranges are in the next version.)

[Squirrelloid]

Please tell me someone has fed Klingon to this.

[Untelligent]

Can anyone recommend a good place for getting language sources (particularily real-world languages, but fantasy languages are OK too)? I could've sworn I saw one the first time I read this thread long ago in ye olde January, but I can't seem to find such a reference, so maybe I'm wrong.

[Igfig]

I get most of mine from Chris Pound's site.

[Nobbins]

Hey, I'm getting a problem whenever I try to use this (Mac, by the way.)

Code: [Select]

python ~/Desktop/df_osx31.10/DFLang/DFLang.py --min=4 --max=9 ~/Desktop/df_osx31.10/DFLang/languages/hebrew.txt

//Yields...

Generating language...
Error opening language_words.txt.
Generation cancelled.


[Igfig]

While investigating the problem, I came across a small bug in the defaults-loading code.  Version 1.8 has been uploaded.

Are you running DFLang from the DFLang folder?  If not, it's probably reading defaults.txt, seeing that the words file is "./language_words.txt", and looking for the file in whatever directory you're currently in.  Since there's no "~/language_words.txt", it causes an error.

Try adding the DFLang folder to your PATH environment variable.  That should make it look in the right place. If that doesn't work, you might try changing the "." in the first two lines of defaults.txt to the DFLang folder's absolute path.  For you that'd be

~/Desktop/df_osx31.10/DFLang/languages
~/Desktop/df_osx31.10/DFLang
language_words.txt
language_SYM.txt

Let me know how that works out.

[nbonaparte]

It worked beautifully for Quenya. Source text.

[Arkbrik]

I just want to say that I really like this generator. In just a couple hours I made languages for all the races in my GURPS setting. Thanks!!

[Lord Herman]

Wow, this is a great tool! I now have faux-Norse speaking dorfs, old Irish for elves, latin for humans, and old Gothic for goblins. Thanks, Igfig!

[Lysabild]

Download link isn't working for me, anywhere else I can get this?