Comments on 2019-01-27 Digraphs and Name Generation

This needs Markhov chains, or at least and indicator for terminal sounds. Looking at “Ngfo” for example: Sure, many words end with a last consonant like “ng”. But that doesn’t meant you can put “ng” anywhere in a word. So actually what we need to do looks like a Markhov chain based on syllables, I think? I guess you will then never get any variations on “John”. Perhaps we could implement a sort of “word2vec” for characters: we’d find that generally speaking “o” often occurs in the context of “j”, “h” and “n” and perhaps we’d find that “u” often occurs in the same context (”June”) and so if we get the “nearest” characters to “o” given the context of “j”, “h” and “n” we might get back “u”, leading to the “new” word “Juhn”.

– Alex Schroeder 2019-01-30 07:35 UTC


Good to see Perl use.

Ngfo, have to use that one. And make up the dialect of a language that has a silent Ng. 😊

Blue Tyson 2019-01-30 10:12 UTC


Heh. 😊

– Alex Schroeder 2019-01-30 13:01 UTC


Please make sure you contribute only your own work, or work licensed under the GNU Free Documentation License. Note: in order to facilitate peer review and fight vandalism, we will store your IP number for a number of days. See Privacy Policy for more information. See Info for text formatting rules. You can edit this page if you need to fix typos. You can subscribe to updates by email without leaving a comment.

To save this page you must answer this question:

Please say HELLO.