making roko was not a very streamline process. at some point, i kind of forgot i was making an utauloid and they just became a funny little person i could toy around with and make friends with vaguely unrelated deities from satire religions. but don't worry about that
✅ recorded
✅ configured
❌ illustrated
❌ formatted
✅ recorded
✅ configured
❌ illustrated
❌ formatted
very amazing ms paint visualization of taoreta
roko's voice was originally meant to correspond to a character i had already created named taoreta ainimari. toareta's character concept was very similar to that of an utauloid, so i thought she would be a good fit. i immediately jumped into trying to record a vccv english bank. that did not go well. i never finished recording it, but i still have the files from that first bank. i also decided at this point that my voice did not work for taoreta.
my next approach was to record a japanese cv voicebank, after looking for something more accessible. after recording the bank and running a few samples through utau, i started thinking of concepts for a design. i never fully decided on a name for this person, but i believe i used parts for "daigiri roko".
unfortunately, i lost the files for this bank... and decided to just start fresh.
and that's where roko comes from. i wanted them to be the same species as a character i already had. both have parents from eurydice, hence the name of this site.
this was the first time i had ever made an utau voicebank. it was made very early into my utau experience and i kind of jumped the gun. i never finished this one (obviously), but i still use some of the files as an appendix for apricot. roko was not conceptualized at the time.
this one is gone forever (oh no)... luckily, it was pretty similar to apricot, so i didn't lose much. the only real difference was that it was a bit sweeter in tone. roko also wasn't a thing yet
i tried to make another soft vb and it didn't go well. i was sick while it was being recorded, so a lot of the sounds are mispronounced. the "n"s were notably jarring. the design concept i had would eventually be used for mercury.
bad.
this bank was deeper than the others. i wasn't satisfied with it, so i tried again.
same idea, but i just stopped trying after this point. it can be set an octave deeper than typical usts and sounds masculine, to an extent.
the reason roko doesn't have a power vb yet
initially, this was supposed to be a third rendition of sapphire after a bit of a break, but before i started recording i decided to just do it at my normal speaking pitch. prior to this, i had been lifting my voice a smidge to record. this is also part of the reason why cherry sounds so different from roko's other banks.
i never finished this one. the idea was that it would be a multilingual version of blue. prior to this, i had recorded some extra samples for strawberry with the intention that i would make a chinese vb. that never came into fruition.
1 cv casual 2 seaglass
1 jade 2 peridot 3 sulu
official
attempt 2
lime's old official artwork
appendix pitch for blue
1 marble 2 strawberry
peach (append)
appendix for peach i think
1 ametrine 2 watermelon
1 mercury
1 quartz 2 limestone
old official artwork for blue
cherry
cherry's old official artwork
i referenced the artwork for jade