TIP: Use SetParam or VLabeler for your VCV otos. They both have plugins that allow you to adjust the oto values en masse, so if you want all your PRE, OVL, CON and RB values to be specific, this does that. They also both have something called "Preutterance Timing Adjusment Mode" which basically lets you oto something in a single click, and they both allow you to navigate the oto with your arrow keys. They're incredibly useful. On top of this, if someone prefers to name their files in hiragana, SetParam has its own auto-otoer which is MUCH more accurate than Moresampler. But since it pulls its alias names from the filename itself (whereas Moresampler guesses), SetParam only accepts kana-encoded files. EDIT: Also for .yaml files, Notepad++ will work just fine.
Hii a jinriki utau maker here!! TYSM FOR THIS TUTORIAL!! I’m trying to find a actual good one with detail since I want to tweak and improve mine and couldn’t find any so yeah.
thanks for the tutorial! soon i'm going to make an utau jinriki vb of my favorite character from leapfrog using my laggy ass laptop (but i'm working on it soon), thank you for this!
Thank you for this! I am finding this very comprehensive so far and I am using your tutorial for a jinriki utau who was originally from a video game (going through his voice lines to find the phenomes I needed was a pain because spoken English is a bitch sometimes when making utaus)
If theres anyone here who's new to making jinriki voicebanks, I'd highly recommend using a 2 mora CVVC reclist instead of converting it to VCV. It takes way quicker, and while there is a small learning curve with otoing, it's easier and still produces the same desired result. However, depending on the quality of the samples, it may sound distorted in UTAU, so I'd recommend using OpenUtau for CVVC jinrikis :)
Yes, what they said^^ you can make a voicebank but the quality will not be as good compared to using singing samples. Also Proseka TTS can't make some specific samples like Ha, He, Hi, Ho, Hu, Wa, Wu, etc. because in Japanese, they don't pronounce it like that.
If you're trying to get characters who don't sing, do you just look up their voice lines instead? For example if I wanted to make a VB of Thomas from Compass (which I do plan to, I'm going to make him sing Metamo Reborn and it will SUCK) I'd just look up a video of his voicelines instead of going through the whole audio remover step?
why i want this (make utaus of artists :^) also if there are english artists that you want to sing in japanese look up how do english people pronounce (,ka,ki,ku,ke.ko for example) or cut and combine parts if possible or if it sounds good or if it doesnt sound good there is always editing the files to make it cleaner/smoothers
STOOOP WHEN I JUST DOWNLOADED OPENUTAU I THOUGHT OF MAKING A VOICEBANK USING MAFUYU'S VA VOICE CUZ I LIKE HER VOICE BUT THOUGHT I WOULD GET CALLED OUT ON THAT 😔😔
On the Vocal Remover part, when I put all the settings and click to convert, it just finishes instantly. I tried to just convert one track, and it has an error that the "Vocal_HP_4BAND_3090_arch-124m.pth" Isn't on the models folder. Did i do something wrong?? I don't know how to redo everything if that is the case
for the vcv reclist are va_va_vi_va_vu_va, va_ve_va_vo_va, vi_vi_vu_vi_ve_vi, vi_vo_vi, vu_vu_ve_vu_vo_vu, n_vu, ve_ve_vo_ve_n_ve, vo_vo_n_vo, n_va and n_vi (all the ones with v) optional?
hi, i want to ask why i can't see my files in the separated folder after i intefered? i even put in the code but that doesnt work either, I'm pretty sure I completed all the steps perfectly
do you have any samples of male voices you could use for crossfading? all but one of the jinrikis i wish to make are of masculine voices so crossfading with a feminine voice would sound Weird Probably
I might be dumb for just blind but everytime i try to extract the audio, it wouldnt go and then i realized that i dont the model the 'Vocal_HP_BAND" one. is this normal or did i mess with the files?
I think there's mistake in your vid. you say to export should be ctrl + shift + r but when I tried it didn't work but when I press ctrl + shift + e instead it worked?
ctrl+shift+R is my own setting, you can see at 2:12 where I displayed my shortcuts for Audacity. You may have your own shortcut set at ctrl+shift+e, but iirc that shortcut exports all the audio and not just the selected one
yeah Jinriki are mostly for private/personal use only. You can't distribute a voicebank using another persons voice without their consent (hence why this is prohibited in Japan)
@@LunaCookie yes that is fine. But it won't be called a "Jinriki" UTAU anymore since you will provide the voice. It will just become a regular UTAU voicebank
Just a quick question, I recently made a jinriki vb and Im using it in utau synth but when I type certain sylables, the program just crashes, is there a way I could fix this problem?
TIP: Use SetParam or VLabeler for your VCV otos. They both have plugins that allow you to adjust the oto values en masse, so if you want all your PRE, OVL, CON and RB values to be specific, this does that. They also both have something called "Preutterance Timing Adjusment Mode" which basically lets you oto something in a single click, and they both allow you to navigate the oto with your arrow keys. They're incredibly useful.
On top of this, if someone prefers to name their files in hiragana, SetParam has its own auto-otoer which is MUCH more accurate than Moresampler. But since it pulls its alias names from the filename itself (whereas Moresampler guesses), SetParam only accepts kana-encoded files.
EDIT: Also for .yaml files, Notepad++ will work just fine.
thanks! I've never used SetParam cuz it seems too complicated for me (lmao)
@@rippi-chan PaintedCz has a really good tutorial on it that explains everything you need to know!
@@rippi-chan so good that you can even make Boyfriend Keith Hatsune (yes,both of them *Miku and Keith* are sibblings) Jinriki
I've already knew that making an UTAU voicebank was hard. But never knew it would be this complicated...
Thanks for the tutorial!!
Hii a jinriki utau maker here!!
TYSM FOR THIS TUTORIAL!! I’m trying to find a actual good one with detail since I want to tweak and improve mine and couldn’t find any so yeah.
TIME TO TROLL MY FRIENDS. Not to extremely
I am a beginner utau user, this helped me a lot and helped me make my first jinriki ^_^
Oh my god this was so useful my mizuki sounds so amazing 😭 THANK YOU
Thank you! This is extremely helpful and gives lots of shortcuts and hints that otherwise would take lots of trial and error to get right.
I did not know it was possible to create VCV from CV until today! Thank you very much for this wonderful info 🙏
THANKS FOR THE TUTORIAL!!! I'm going to make Tsukasa>:3
thanks for the tutorial!
soon i'm going to make an utau jinriki vb of my favorite character from leapfrog using my laggy ass laptop (but i'm working on it soon), thank you for this!
Thank you for this! I am finding this very comprehensive so far and I am using your tutorial for a jinriki utau who was originally from a video game (going through his voice lines to find the phenomes I needed was a pain because spoken English is a bitch sometimes when making utaus)
If theres anyone here who's new to making jinriki voicebanks, I'd highly recommend using a 2 mora CVVC reclist instead of converting it to VCV. It takes way quicker, and while there is a small learning curve with otoing, it's easier and still produces the same desired result. However, depending on the quality of the samples, it may sound distorted in UTAU, so I'd recommend using OpenUtau for CVVC jinrikis :)
I remember a site where you could put in letters and the project sekai characters would say them. Do you think that could work for the Jinriki UTAUs?
The site is called Project Sekai TTS so maybe?
Yes, what they said^^ you can make a voicebank but the quality will not be as good compared to using singing samples. Also Proseka TTS can't make some specific samples like Ha, He, Hi, Ho, Hu, Wa, Wu, etc. because in Japanese, they don't pronounce it like that.
If you're trying to get characters who don't sing, do you just look up their voice lines instead? For example if I wanted to make a VB of Thomas from Compass (which I do plan to, I'm going to make him sing Metamo Reborn and it will SUCK) I'd just look up a video of his voicelines instead of going through the whole audio remover step?
Yes you can use talking audio too!
for when the full version of Chase with kanade 😭✨
I will be using this to make fan covers possibly fansongs but mainly fan cover
This guide is so useful, thank you so much!
why i want this
(make utaus of artists :^)
also if there are english artists that you want to sing in japanese
look up how do english people pronounce (,ka,ki,ku,ke.ko for example)
or cut and combine parts if possible or if it sounds good or if it doesnt sound good there is always editing the files to make it cleaner/smoothers
thank you i am going to make an absolute monstrosity with this
THANK YOU FOR THIS WAAAA I always wanted to know what it was like in audacity to easily trim the audio
my laptop has so many files now after this but very usefull
Thank you for the tutorial i think it will be useful for otoMAD
STOOOP WHEN I JUST DOWNLOADED OPENUTAU I THOUGHT OF MAKING A VOICEBANK USING MAFUYU'S VA VOICE CUZ I LIKE HER VOICE BUT THOUGHT I WOULD GET CALLED OUT ON THAT 😔😔
Sorry what song kanade sing? 0:24
chase - setsuna yuki iirc
On the Vocal Remover part, when I put all the settings and click to convert, it just finishes instantly. I tried to just convert one track, and it has an error that the "Vocal_HP_4BAND_3090_arch-124m.pth" Isn't on the models folder. Did i do something wrong?? I don't know how to redo everything if that is the case
I just have one question. What does tuning mean? Is it that we record our singin and switch it with our chosen voicebank?
Have you considered making jinrikis from SekaiTTS?
Yes I did! I tried tsukasa and it was really bad 😭 it seems his talking voice is very different from his singing voice lol I'm gonna try emu next
thx now i made my friend's oc
Hi! Where can i download the rui cv? Also are there any regular utau tsukasas?
Hi! I don't think anyone has released a Jinriki voicebank of project sekai characters. I don't release any of my voicebanks too ^^
for the vcv reclist are va_va_vi_va_vu_va, va_ve_va_vo_va, vi_vi_vu_vi_ve_vi, vi_vo_vi, vu_vu_ve_vu_vo_vu, n_vu, ve_ve_vo_ve_n_ve, vo_vo_n_vo, n_va and n_vi (all the ones with v) optional?
10:09 Friday night funkin but music is faster be like
hi, i want to ask why i can't see my files in the separated folder after i intefered? i even put in the code but that doesnt work either, I'm pretty sure I completed all the steps perfectly
13:36 YOU HAVE TOUYA SCREAMO????
do you have any samples of male voices you could use for crossfading? all but one of the jinrikis i wish to make are of masculine voices so crossfading with a feminine voice would sound Weird Probably
if you mean the consonants, I've uploaded one in my resource folder which are sourced from akito and toya^^
@@rippi-chan okay!
I might be dumb for just blind but everytime i try to extract the audio, it wouldnt go and then i realized that i dont the model the 'Vocal_HP_BAND" one. is this normal or did i mess with the files?
ok nvm i got the model AHAHAHAHAHAA
WHERE DID YOU FIND IT?? For me it's the same, idk if i installed it wrong or something
do u have any song recommendations for making a rui jinriki because im struggling finding one
hi so im using audacity like you said and 'Ctrl + Shift + R' dosent work. what do i do
is it possible to OTO in OpenUTAU?
very good tutorial btw :)
I think there's mistake in your vid. you say to export should be ctrl + shift + r but when I tried it didn't work but when I press ctrl + shift + e instead it worked?
ctrl+shift+R is my own setting, you can see at 2:12 where I displayed my shortcuts for Audacity. You may have your own shortcut set at ctrl+shift+e, but iirc that shortcut exports all the audio and not just the selected one
Are Jinriki UTAUs in private?
yeah Jinriki are mostly for private/personal use only. You can't distribute a voicebank using another persons voice without their consent (hence why this is prohibited in Japan)
@@rippi-chan Okay. Is it fine if I provide the voice of my OCs (like my Yu-Gi-Oh! protagonist OC Yuka Aiuchi) before putting them into UTAU?
@@LunaCookie yes that is fine. But it won't be called a "Jinriki" UTAU anymore since you will provide the voice. It will just become a regular UTAU voicebank
Can you send me a converted .aup file instead of .aup3? I don't wanna use Audacity 3.0 since it's basically the same thing as ever
Just a quick question, I recently made a jinriki vb and Im using it in utau synth but when I type certain sylables, the program just crashes, is there a way I could fix this problem?
what kind of syllables? is it in hiragana or romaji?
@@rippi-chan it was a hirigana cv voicebank
LMAO THIS ALSO HAPPENS TO ME 😭😭
I think the problem is 'cuz you don't have all the phonemes??? If not, then I have no idea, I'm wondering that too.
what is the difference between utau and open utau
hello, where i can download the emu cv voicebank?
I use akio from wangan midnight series
i made a rui english cvvc bank
what do i do if it sounds extremely echo-y
the samples? then you need to clean it up by removing the echoed parts