You now have a couple of choices: either Create a regular-expression lexicon file called monish-lex.regex, compile it using read regex monish-lex.regex and save stack the result as monish-lex.fst; or Define the lexicon using a script file and run it using the source command. Thisscript file should save the final result as monish-lex.fst. Either way, the lexicon file will describe the morphotactics of Monish words, and the resulting network should have baseforms and tags on the upper side, and baseformsfollowed by abstractsuffixes on the lowerside. Assuming that you choose to write your grammar as an xfst script, you will have definitions like the following: define Root [ r u u z o d | t s a r l o k | n t o n o l | b u n o o t s | v e s i i m b | y a a q i n | f e s e e n g ] ; define Suff1 %+Int .x. [ %U %U k ] ; define Suff2 [ %+Perf .x. [ %O n ] ] | [ %+Imperf .x. [ %O m b ] ] | [ %+Opt .x. [ %U d d ] ] ;

2. You now have a couple of choices: either - Create a regular-expression lexicon file called mon ish-lex. regex, compile it using read regex