First, set up your recording area. You're going to desire a quiet area without a most echo, a decent mic (I've watched MXL USB mics and also Blue Snowballs/Blue Yetis as being nice popular), and probably Oremo on her computer. Oremo is set up to instantly oto as well as carry out background music for you to document to, for this reason you store a good pitch and also a steady beat through your samples.
You are watching: How to make your own utau
Once you have actually a setup, you require a reclist, or recording list. This is a list of the sound you need. Because that English, there are actually a couple of different methods. There's CVVC (consonant-vowel-vowel-consonant), which is ns think the very first style the English UTAU to be developed. I can be wrong. The is thought about somewhat dated, but if you desire to use this layout there's a reclist accessible in this thread. The very first link is broken, however there's a reupload down a few posts. If CVVC isn't her style, there's VCCV (vowel-consonant-consonant-vowel), i beg your pardon is much more standard now. There's a tutorial collection by the reclist maker, PaintedCz, here. This is the layout most English USTs are in. So if friend wanna cover the cool brand-new VocaCircus song and you snag a ust because that it, it's probably in VCCV format.
Finally, if those aren't your style, there's an even newer technique called Arpasing, which supplies a totally different system for writing the sound (for instance,
in VCCV i do not care "ae" in Arpasing). There's a website at https://arpasing.neocities.org/ which has test USTs, a reclist, and also a pronunciation guide for phonemes, and also more. The developer is Kanru Hua, who additionally made the UTAU resampler/wavtool/voicebank configure thingy Moresampler. If friend want, you deserve to actually plug a CV, VCV, or Arpasing voicebank into Moresampler and the routine will provide you a an easy oto, but I'd recommend tweaking it.
Once you select your recording room and voicebank style, it's time come record. This is gonna take a while.
Configure the sounds. When Oremo (and Moresampler for an Arpasing VB) do this because that you, it generally isn't perfect. You can skip this step, however again, fine tuning never hurts.
See more: Driving Distance From Cincinnati Ohio To Chicago Illinois, It'S 297 Miles From Cincinnati To Chicago
After that's done, test your voicebank with a ust. Give it a name. There room a couple other things, but the bulk of the is recording a shitton the sounds and also making sure the routine knows what components of the sounds room what. Maybe making a personality to go v the voice.
There are discord servers to help you out as well, and also stuff ~ above utaforum. Expect this helped