First, list all vowel *phonemes* the language has.
Then list all single consonant sounds that language has.
And lastly check that what of those consonants can act as an ending consonant like “t” in English word “cat”.
If the language has possible consonant+consonant combinations, write them down too (e.g. “tr” in “tree” or “st” in “test”).
Ok, now you know the all phonemes and phoneme combination that you’ll need.
The easiest pattern to write reclist is CVCV(C). Put ending C if that is possible, other wise just CVCV.
For example “haha” and “kakak”. From these samples, you’ll able to get [- CV] [V C] [CV] (and [V C-] from k sample).
When all C’s are paired with “a” (babab, cacac, dadad... zazaz), then take the next vowel phoneme and reply the same.
The basics are covered, let’s now cover CCs.
Ask yourself, is this CC appearing beginning, and/or end of word.
If in beginning write as CCVCCV. If in end only then as VCCVCC. If it can occur in both beginning and end of a word then as CCVCCVCC.
For example, “st” can appear in English both beginning and the end of word like in “star” and “just”. So CC recordings for this sound will look like ”stastast”. From this, you’ll get [- CCV] [V CC] [CC V] [V CC-] of “st”.
OR
If you’re lazy to record, you can also just configure just the consonants and not vowels like [- CC] [CC] [CC -]. You can take a reference of this from for example Cz’s VCCV English and Delta’s ENG CVVC.
I recommending configuring CC only if the target language has an insane amount of CC that would take otherwise ages to record.
Good luck!