Best Way To Do An English Voicebank?

I’m considering porting/switching some of my UTAUloids over to DeepVocal, as I very much like the interface and parameter editing. However, I’m struggling to figure out the best way to approach English. I think that recording unique CV samples for every consonant cluster I want is probably the best way to go, even if it means having more samples. But ending consonants have me a bit stumped.

The way I understand it, the only way to do ending consonants is via the tail symbols, but this prevents any CC or CV connections to following words. Is that just how English in DeepVocal works right now, or is there another way? Can ending consonants be programmed as a CV or will DeepVocal give me an error when building the model if I try that?

i didnt make a voicebank before but the best i can think is “cvc” but the thing is you had to record like “bab, bad, bak etc” it will be a lot so i dont recomend, but diphthongs are easyer in deepvocal


and thats why i think cvc is the best way for deepvocal 1.1.6

“im trying to import utau english vb to deepvocal bc im also trying what is the best english style for deepvocal”