I'm a retard can somebody explain to me the difference between VCV, CVVC and VCCV? I understand that CV is the typical robotic sounding voicebank.

>>511
C = Consonant
V = Vowel

These are called "reclists" and they're basically just how the voicebank is recorded. Adding Cs and Vs is supposed to make the bank more smooth/realistic but I can't really hear the difference (in Japanese) past VCV (I don't listen to CVVC/VCCV banks often though and I definetely don't have the skills to use them myself) although CVVC and VCCV are super useful for languages that aren't Japanese, especially considering CV English isn't really possible/would sound VERY bad.
It can go all the way up to VCVVC lol, there's also C+V, CV+VV, CVVChinese, ARPAsing (which was made by the guy who eventually went on to create SynthV) etc.
I've heard that "C+V English" is a thing but I've never found a reclist.

>>512
Which of these is usually recommended for smoother results? I only have experience making CVs

>>513
VCV for Japanese, usually. Unless you have interest in making a bank in a different language.

Can you stretch out a syllable in UTAU like in Vocaloid/SynthV where you type "-"? I'm trying to make a Rei cover of RIP=RELEASE but the "A A A A" part sounds choppy and weird.

>>529
Possible in OpenUTAU (with + instead of - though iirc), al though not in original UTAU. I usually just duplicate the vowel (ex. [ma a] instead of [ma -])

testing.mp4 - 1122.78 KB (636x358)

werkz

should i go with tunelab or openutau

Which resampler works best with breathy vocals? Moresampler is typically my go-to but its a bit meh in that regard

Genuine question, where do you guys find people to voice your UTAUs? Or do you voice them yourselves?

>>1517
usually myself or family members that are willing
is anyone running original utau on linux? i got it working once but the rendering was fucked, too stubborn to learn openutau though

>>1520
>too stubborn to learn openutau
It's literally easier than 2008 Utau

>>1523
ya I was originally also too stubborn to learn but then when I did I was like. oh. this is easier. wtf. so ya I 100% recommend learning it lol

>>1523
>>1524
i guess it wouldn't hurt to try it out again
i stopped using it at first because it sounded really weird with my vbs but there's probably a way to fix that

>>1525
The default resampler kinda sucks, that's probably why. Get moresampler, its great.

>>1525
update. yeah no i'm a fucking idiot this is easy as shit i can see why so many people use it now

ava_crop.png - 116.25 KB (555x555)

Just finished A.V.A-4.0 (Adaptive Voice Assistant)'s voicebank, an inexpressive robotic CV voicebank that still sounds human enough. This is probably the first voicebank to be released here. An anon from here voiced it and i created the character's original design (which doesn't look much different from the final one, i think the major difference is that her boobs are bigger wwww).

I'll post the download here but i need to let you all know that it's a bit buggy. Not sure if the issue is the otoing or the resampler but some phonemes don't work. How 2 fix? I DONT KNOW

Stuff no one cares about:
>Height: 157cm(5'1)
>Likes: Helping
>Dislikes: Being useless
>Personality: Quiet
>Interest(s): Music
>She is a robot assistant whose job is to help producers make music whenever they don't have a vocalist available. Her eyes are LED screens which light up when on, however, they are useless, it's the camera on her head allows her to see.

https://files.catbox.moe/e8cmio.zip

lazycrazyd….png - 503.41 KB (2735x4298)

>>1599
Original design for comparison, it's a bit more toony

Are there any tools that you can oto voicebanks with? OpenUTAU's one sucks
>>1599
Cutie patootie. Will definitely DL

>>1602
There's vLabeler and setParam but i just use UTAU's (Not OpenUTAU) default otoing tool

>>1599
I can't get the ba,be,bi,etc sounds to work

>>1604
Yeah no clue either. I didn't oto her, i just edited her voice and did her design.

Does anybody know how to convert a CVVC UST to CV and a CV UST to VCV?

>>1734
use utaformatix
for the second one if you use openutau you can click 'default' in the singer tab, go to 'japanese' and click vcv, it'll do it automatically

Does anybody have an ECHO ust that works with japanese voicebanks?

>>1940
Oh, I think I used to have one but I’m not sure if I still do. I remember finding it by looking up engrish in the search, though.

Oremo or Audacity for recording?

>>2024
Oremo for CV, Audacity for anything else

ClipboardI….png - 5.36 KB (1070x66)

Any hifiresampler users in here? It just came out a few days ago. I've been trying to set it up and followed the entire guide but got stuck at run.bat which comes up with the following message. I believe i have torch installed already since attempting to install it manually via cmd tells me it's already installed.

2025-03-20….mp4 - 3039.51 KB (1920x1080)

This Hifisampler shit is really good

>>2372
this shit is like crack fr
I installed it today after hearing this and did this
https://files.catbox.moe/duxbeh.mp3
there's more to it but I'm saving that for public later

>>2385
disclaimer: I've never actually done crack so idk if it actually is like crack
but
I just spent all day playing with hifisampler so maybe?

>>2385
funny how SV got worse and UTAU got better the same month

i made my own bootleg adachi rei by partially using her method

https://vocaroo.com/14LXbHRSmNRW [Embed]

Play this Vocaroo recording

Does anyone have the Ring Suzune UTAU VB people made from her vocaloid samples?

>>2437
I don't have it but i do have the V3 port someone made. It's not UTAU but i guess it's still something.
https://archive.org/details/vfs-01r-ring-suzune-r1

>>2385
this sounds better than a lot of V4s

Screenshot….jpg - 119.70 KB (1919x259)

bro why did my hifisampler kill itself (dw my laptop name has nothing to do with me)

>>2437
Oh I found it
https://archive.org/details/ring-suzune-vcv

>>2569
I literally just turned my pc off and on again and now it works, why does the default IT fix always works

2ZiJo2h0VD….jpg - 107.46 KB (750x509)

>>1599
>>1601

hello! it's me, AVA's (Adaptive Vocal Assistant) voice provider. i have decided to correct AVA's oto.ini for easier and simpler use.

give this unofficial leekaloid(?) some love!

VB with oto.ini fix download: https://files.catbox.moe/bp25gf.zip

voice sample:

https://vocaroo.com/17H9kJnZIhwM [Embed]

Play this Vocaroo recording

>>2676
small fix (AGAIN), i forgot to fix the "u" phoneme lol
https://files.catbox.moe/gul2hs.zip

>>2676
I just realized this means TORIKO of all people is the most “normal” of the 3 leekaloid so far since she’s the only human

>>2679
at the cost of her mental health i guess

I would consider AVA a honorary leekaloid. She wasnt made by a team but she was the first UTAU to be released here.

ClipboardI….png - 76.96 KB (223x250)

>>2679
she's only human after all

How do you access the open beta features in openutau

>>2679
>Normal

Can someone suggest an UTAU i can use for English songs? Preferably one that doesn't look like it was made by a Deviantart user in 2011

>>4153
Unironically teto
Her english voicebank is fine

>>4157
toriko diffsinger :(

OpenUTAU bros how do i achieve this slight rasp on the "te" and "ko" at the beginning? I've heard Slave VVR do similar things with Miku and Giga has also done this with his recent song.

https://youtube.com/watch?v=423Nmfpo828 [Embed]

https://youtube.com/watch?v=7z80uGHJyjA

>>5502
It's the vocal fry, here's a tutorial

>>5503
Thanks

https://youtube.com/watch?v=7uxi6TXijd8

inanimate horse utau

>>5927
oh god they made an utau out of this... i'm not even mad, he looks and sounds cute

>>5927
This entire thing's fandom is just people who are unaware that Uma Musume exists

>>5953
They probably know

image_2025….png - 136.92 KB (600x457)

Hello everyone! As a data hoarder, while roaming my external hard drive, I found a copy of MikuMikuVoice.
https://files.catbox.moe/ps4a6j.zip

Here's what VocaDB says about it:

MikuMikuVoice is a freeware plugin published on May 29, 2008 that automatically generates .VSQ files and pitchbends from vocal audio. It can be used for both music and speech vocals. Download links were discontinued after the release of VocaListener.

>>6093
I've also mirrored it to archive dot org

https://archive.org/details/miku-miku-voice_202506

Vocal fry test in OpenUTAU with Sena.

https://vocaroo.com/1cSQZKcKASlh [Embed]

Play this Vocaroo recording

>>6335
It bothers me that Sena is the only LEEKAloid that has breath samples

>>2427
This is essentially Shion

ClipboardI….png - 20.10 KB (891x405)

How the hell do you get CV's working on TuneLab?

>>6644
Nevermind got them working. Seems to be only Shinya that doesn't work for whatever reason.

20250707_1….jpg - 145.16 KB (900x1200)

This design caught my eye earlier

https://x.com/utau_catalog/status/1942067218998526180 [Embed]

Show the video in this Twitter/X post

GtkAUUoX0….jpeg - 428.77 KB (2602x4100)

>>7329
Found this account from this tweet and it seems like they're a database of UTAUs that you can contribute to (at least that's what the link in bio seems to imply). I've never seen anything like this before so this is neat.
https://x.com/utau_catalog

Check this guy out, he's adorable...

GvPSxpZWQ….jpeg - 122.64 KB (900x1200)

>>7344
Doll girl utau
https://bowlroll.net/file/327306

https://youtube.com/watch?v=I5DBNqInzaI

New song from a relatively unknown producer. I like it

cielpo.png - 693.92 KB (1000x1800)

Finished my awesome visual kei utauloid I made because I was bored and drunk one day. An anon here designed her and I provided her voice and initial concept.

Here's some dumb shit:
>CIELPOID is an emo vkei idol hailing from the streets of Shibuya. Said to have miraculously crawled from a broken mirror, she feigns a confident and mystical personality in public, but spends most of her days sitting in her apartment building practicing Rider-esque attack calls and watching toku reruns she taped 10 years ago with her cat. Often too busy sleeping to notice the debt she's in from how many magical girl figures and mecha model kits she impulse buys.
>CIELPOID is based on the voice of utaite and vocaloid producer RUINxCIEL/shieru-p.
>Age: 20
>Height: 185cm(6'1)
>Gender: Soft Visual
>Character Item: Spray Painted Desert Eagle (x2)
>Likes: Playing pretend with Tokusatsu belt toys, dead MMOs, curry, painted fingernails (fetish), strawberry milk, mecha, 「ボーイズラブ」, Kamen Rider Ryuki
>Dislikes: Deadlines, toothpicks, debt collectors
Preview:

https://vocaroo.com/15sapjeHVmdY [Embed]

Play this Vocaroo recording

https://vocaroo.com/1n0k792c1g3J [Embed]

Play this Vocaroo recording

https://vocaroo.com/1g9Fg0zVSX7q [Embed]

Play this Vocaroo recording

Download:
https://files.catbox.moe/yld4qb.rar

>>7524
the voice reminds me of male ruko voicebank

>>7527
initially she was going to sound more feminine, but I thought it'd be amusing (and moe) to give her a guy voice

>>7524
I love her lil ahoge it's so cute

Does anybody have English VCV reclists

>>7524
shieru Where did you get the idea to give her leg warmers? I wanna get some for going out they’re so cute I always loved leg warmers

https://youtube.com/watch?v=G5x94yWTRUc

>>7625
It might technically be CVVC, but the Arpasing reclist has gained some traction among English UTAU. Here’s a list of several different variants.
https://arpasing.tubs.wtf/en/directories/reclists
Personally I think this recording style has more intuitive phonemes and results in more natural-sounding English than previous recording methods. You might also want to check the details on OpenUTAU’s integration to make things more seamsless for users

d6440aa050….jpg - 87.30 KB (720x1562)

>>7629
You say that like Vocaloids with leg warmers have never been done before

IMG_9291.jpeg - 36.92 KB (378x750)

>>7641
Somehow I never registered that Rin’s leg warmers are leg warmers.. I just think of Amu when I think of leg warmers.

>>7629
Her design was partially inspired by the Kagamine twins, shoujo manga and older fanloids and I noticed that was something common with certain designs of those. Also they're just cute

>>7524
she is now on the TCC it looks like!

>>7524
But does she pass the high range test?
(UST: Lasaillax)

https://vocaroo.com/1boi4QdXOhPO [Embed]

Play this Vocaroo recording

>>8029
Does Sena even pass the high range test? I don't think she does

>>8029
for a monopitch vb she surprisingly sounds really good at higher ranges. the advantages of having a lower pitch, i guess!

>>8029
shockingly so haha
>>8035
i recorded her lines at around e6, which was probably my most comfortable range for hardcore/emo vocals specfically, so i was honestly kind of shocked at how well she sings at both much lower and much higher ranges

here's some silliness i recorded today:

https://vocaroo.com/1kr5srZf06bt [Embed]

Play this Vocaroo recording

Has anyone used DeepVocal, possibly even converted an UTAU bank to a DeepVocal one? I assume it’s worse than diffsinger but I heard the quality is a bit better than Utau, but I assume with HifiSampler that may not necessarily be the case anymore either…

>>8099
There is a tool to convert them but i didnt quite understand how to use it

>>8101
I see, I’m more so just curious about the different levels of quality though!

>>8106
It's slightly better but UTAU VCVs with Hifisampler sound even better

>>8110
Since UTAU VCVs with Hifisampler are said to sound like V4, what does DeepVocal sound like? V2? V3?

https://youtube.com/watch?v=9q-QLtYX3cQ

>>8117
I guess you could say that

cielpoidjbf.mp4 - 4119.61 KB (1920x1080)

bored so messing with cielpo again, just be friends UST by landy25n
might see if i can get her to sing in english later

>>8317
aren't you the guy that made frail

>>8318
unfortunately

>>8320
>unfortunately
It was a good song brother

Is moresampler still *the* best resampler when it comes to flags and overall sound quality?

>>8386
It still has the most flag options, but Hifisampler is the new and cool HQ sampler.

>>8386
It’s still very good. I try out multiple resamplers per voicebank to find the right one for it. Moresampler and hifisampler are my favorites though.

>>8386
Moresampler trashes the quality of half my VBs, but has the most flags and sounds great on the ones it works on.
Hifisampler is the best in terms of quality, but it relies on a local AI server so it uses a lot of RAM and power, and has no flags. It also cannot run Ruko M Kire or Sukone Tei (and most VCV VBs, but Teto works) but is my absolute favorite.

>>8392
>Sukone Tei
Yes it can? See >>2372

>>8393
Yeah idk what Anon is talking about, both people in that thread were using Tei to test hifisampler

>>8392
I use a portable version of hifisampler and not only does it render as fast as moresampler but it also works with VCV's. Have not tried Ruko but VCV Tei definitely works.

>>8393
>>8402
Thanks, I know on my end she gives errors here and there. Most aren't anything to worry about, but occasionally one will mess up the render. Nothing's as bad as CASE on my end regarding Hifisamp, where certain phonemes fail to load at certain lengths. I should probably update HS since I'm on the second latest version iirc.

momo.png - 6668.09 KB (2515x4000)

Any clue where I can find a download for this UTAU? Her name is Momo Nene-ume

A mobile port of OpenUTAU is apparently being worked on
https://github.com/vocoder712/OpenUtauMobile

>>9635
https://utau.fandom.com/pl/wiki/Momo_Nene
Scroll down and there's a dl link for a VCV bank

what's everybody's go-to resampler these days? still moresampler?

>>10813
I almost never use Moresampler admittedly. I use straycat-rs because it works with just about every bank with the only "issue" being a more unrealistic sound, and hifisampler 0.0.3 for voicebanks that don't hate it (like mine) when I don't mind a ridiculous rendering time or running a script separately.

>>10814
Hifisampler is good if you're willing to sacrifice parameters such as growl, gender, etc. I never had slow rendering issues with it (not the portable version at least) but yeah it lacks a lot of features.

I'm bored and I haven't been here in a while (health problems, burn-out, romantic despair and focus on writing non-vocaloid music, please forgive me) so here's Bad End 39ght. You don't know true suffering until you've tried to mix 8 UTAUloids wwwwwwwwwwwwwwwwwwww
Sena - Miku
CIELPOID - Gakupo
Toriko - Rin
Shinya - Len
Sena (low) - GUMI
A.V.A. - Luka
Akiha - MEIKO
CIELPOID (low) - Kaito

https://vocaroo.com/1doAXhaUz1cz [Embed]

Play this Vocaroo recording

>>10813
wordliner is the only resampler that really sucks tbh

>>10817
this sounds insanely good

>>10818
phaavoco would like to have a word with you.

>>10815
It's my favorite resampler. I've heard 0.0.6 adds flags, but the setup is entirely different and I can't figure it out for the life of me.
>>10818
Doppeltier sounds like shit with 98% of voicebanks, too.

>>10823
There's a new version of hfs? Link it please

Can '26 be the year of UTAU? Bring back robotic voicebanks and wholoids

>>12070
There's definitely more of an interest on UTAU these days. Although its more of an interest on the voicebanks than the actual software. People use Nurse Robot, Teto, Rei, etc but with TT making the jump from UTAU > V6, Teto from UTAU > SV and more recently Rei from UTAU > V6 there's going to be less of a reason for producers to use UTAU.

>>12070
It's always the year of UTAU anon, we just have to spread more UTAU propaganda

>>12096
Better pray Iyowa picks it up again, he's the reason Rei blew up

I wanna learn OpenUTAU. Is there a certain voice bank anyone would like me to try? Cuz I'll make stuff with whoever I use. Only thing is I wanna learn an English CVVC/VCCV or something similar cuz I like those

>>13980
easiest to use is teto

>>13980
openutau has an en -> jp phonemizer (basically lets you type english words and it'll convert it into japanese syllables for japanese utaus) and it works with pretty much any kind of jpn voicebank. so really you could use anyone, and it'll sound fantastic

>>13981
I knew she was the most popular, but I didn't realize she's the easiest, that cool! I was wondering if anyone had an smaller fav that doesnt get much love that I could try tho
>>13983
Oh cool! I'll keep that in mind!

>>13985
>>10536

>>13985
To be fair all utaus are easy to use if you know what you're doing but if you're a beginner i guess Teto would be a good starting point since she's babby's first UTAU voicebank.

>>13980
use one of the leekas, they don't get original songs and barely any covers

has anyone ever made an UTAU where every soundbite is a different voice?

>>19005
i've never seen that. maybe that's something we should do (even if it sounds bad, lel)

https://youtube.com/watch?v=FbVcNiGIxUM

>>19006
donka fjord is, quite literally that lol

>>19021
Am I crazy or one of the samples is literally Mama Luigi

recorded an utau for the first time in 10 years lol i have no ideas for her design or anything and just did it for nostalgia

fjsdko.png - 13.24 KB (480x586)

>>20870
you can have this design i made months ago, i call it fjsdko

>>20871
>>20870
fjsdko has leeka potential

>>20871
omg i love her...

>>20871
The leeka to end all leekas

ClipboardI….png - 17.96 KB (483x479)

>>20874
>>20875
>>20876
she says thank you

mtxz53.png - 1264.54 KB (3072x3072)

>>20877
silly creature i love her uuu

>>20870
Anon you gotta go with this design: >>20877 >>20886

>>20890
if the artist doesnt mind i actually love it so much!! i gotta fix her oto tho i kinda forgot how to do it.....

>>20894
yeah go for it

>>20870
release her

>>20886
I thought she was squatting here
Which gave me an idea, someone should redesign her to be more frog-like

fjsdko fro….png - 14.96 KB (489x422)

>>20932
froge

>>20959
If we make this into an actual leeka we should make her age 10. Not only because she looks young but also because in frog years that means she's very old and on the verge of death

>>20961
don't kill froge

asdkashjka….png - 48.50 KB (574x623)

https://files.catbox.moe/3t5ij8.zip
i kinda fixed her oto?? not the best but it works (ᵕ—ᴗ—)
shes a monopitch cvvc for now!! she sounds kinda fart cuz of my mic too, i'm sorry for that...

>>20959
Remake it with smoother lineart and its perfect

>>20959
my daughter fujisu deco looks so beautiful here

>>20981
Fujisu Deco?

>>20982
sounds too much like the company imo

i like fjsdko better because it sounds exactly like how she looks

ehe.gif - 435.30 KB (310x400)

Your fortune: Godly Luck

>>20974
She has a funny accent

>>20870
Are you Finnish by chance?

fj render.png - 7.42 KB (320x415)

>>20980
we dont need smoother lineart

>>20985
I really like this gif

>>20974
Nice

>>20959
what a beautiful creature!

For whatever reason I'm slowly starting to like the sound of UTAU better than Vocaloid, SV, etc

>>20974
Can we get a VCV?

>>21511
Also why hasn't she been added to TCC

>>20961
Character item should be a Lilypad or a Lotus

>>21513
Lotus could be added to her design maybe

>>21513
<Likes: Ribbit
<Dislikes: Reddit

>>21513
a big lilypad umbrella?

>>21594
Good idea

My suggestion is to make her more stupid looking

froggy vcv is done!!
https://files.catbox.moe/afslve.zip

>>22060
Finally, a VCV leeka...

>>22060
>VCV leekaloid
Can't believe it took this long

>>22060
She’s perfect.

>>22060
cute

Ok so we already defined her age (10) but what about likes, dislikes, etc. What DO frogs like and dislike?. Her character item can be a lotus flower.

The urge to make an ARPAbet VB because I don't want to learn and fight with the diff singer process

>>22100
Are you the frog girl vp?

>>22095
dislikes: fruit, veggies

>>22119
Is that something frogs actually hate

>>22120
yeah, frogs are carnivores

>>22119
>>22120
>>22123
But does she eat insects

Holy shit.
Someone is making an UTAU resampler using the source code that was used for VOCALOID1.

https://x.com/VOISTAROfficial/status/2039890387427365243 [Embed]

Show the video in this Twitter/X post

>>22298
This owns. Wonder how it would handle breathy banks

>>22298
Now Tei can sound even more realistic

>>22298
This is gonna be really good for CVVC and VCV banks

>>22298
Oh fuck, this is awesome news

ClipboardI….png - 220.33 KB (587x503)

there's a faster hifisampler?

>>22303
She already sounds insanely good with Hifisampler, check >>2372

>>22315
This seems to be it
https://github.com/organization/hifisampler-go

Your fortune: Mikudayooo-

>>22323
How do you even install this? Has anyone here tried?

>>22298
Is it supposed to sound like V1 because I'm not hearing it

>>22323
i think they took it down?

>>22752
I still have the exe if you need it

>>22303
Hifisampler is more realistic

Hey /leek/, give me some UTAUs you like that aren't super popular

>>23608
I like KYE a lot. He has a unique tone

>>23608
>Cherilee
Honestly, all of GraySlate's UTAUs are of great quality, but she's my favorite of the bunch.
>Ichiro Honda
My second favorite GreySlate bank
>ALYX
Highly underrated VCCV English bank
>Nukunoko
CV voicebank by Owata Pikosuke. As far as CVs go, I really love it.
>JYOKI
Pairs extremely well with my UTAU (Matsudappoiyo too, but I refuse to accept he's a whoTAU) so I'm biased. Is accented, but I think accented banks are very underrated.
I also like my own UTAU but I've heard from numerous people he's difficult to use, namely because he's extremely picky about resamplers and only seems to sound good with Straycat-rs or Hifisampler, and tolerable with TIPS.

>>23614
Any utau that isnt part of the big 3 is a Whotau sadly

https://youtube.com/watch?v=k0rIwGMlT3c

>>23615
Matsudappoiyo has one (1) cover I love a lot so he isn’t a whotau to me, in my heart.

https://youtube.com/watch?v=HPqOuzuuteU

>>23615
I was about to refute that, then I saw the views on this and realized my introduction to him only has 16k views

>>23618
As much as it sucks the average vocaloid producer doesn't care for UTAU beyond the big 3 (Defo,Rei,Teto). Before Rei it was Momo but idk why she's no longer as popular.

https://youtube.com/watch?v=Kt4j5a0RfZI

>>23617
sounds good
in exchange have another Teto & Matsudappoiyo cover

d2562b420e….jpg - 26.40 KB (500x500)

What's the UTAU equivalent to Moetron

>>24091
We were gonna make it, and then we didn't

ClipboardI….png - 936.19 KB (768x1024)

>>24091
I dont think it has a voicebank but this is what instantly comes to mind

>>24136
Mashing all leekas samples together to create the ultimate leekaloid

>>24098
We really should at some point

>>24136
There is an rvc voicebank of this

08c.png - 419.78 KB (989x1200)

>>23608
my favorite when i was younger was Ariya Gii

>>24248
I like their dino outfit

櫻花アリス.png - 106.61 KB (207x600)

>>23608
Ouka Alice seems a bit forgotten

>>24259
Cute design

hey, so. is it normal for openutau to just not fucking work at all? installed it again to try out Toriko's diff and the Klattsch c+v port but no matter what phonemizer or renderer I try I'm met with fuck all. I would appreciate some help cause this week has been nothing but piles of big, steaming, heaping piles of dog shit.

>>25077
Are you using the Lunai fork or just regular OpenUTAU?. Regular OpenUTAU can't open it and the fork linked in the DiffSinger thread is obsolete.
Download the one here
https://lunaiproject.github.io/editor

>>25078
didn't know lunai existed until now. getting that fixed the issue with lattsch but tori's still busted no matter what phonemizer I use.

ClipboardI….png - 57.28 KB (1035x364)

>>25079
Just to make sure, you are using the phonemizers under More > DiffSinger and not the UTAU ones, right? Because I had the same issue and it turns out I wasn't using the DiffSinger phonemizers

ClipboardI….png - 60.39 KB (614x717)

>>25080
yeah i'm using the diff phonemizers, I'm being greeted with this everytime I press play

>>25083
You're likely missing nsf_hifigan. Go to Tools > Packages and install nsf_hifigan from there.

>>25084
just installed and I'm still getting the same error

>>25086
Post log

ClipboardI….png - 61.92 KB (621x737)

>>25088

>>25089
Idk if this will work but you can have my nsf_hifigan, straight from my Dependencies folder.
https://files.catbox.moe/92v999.zip

>>25090
this fixed it, thank you so much

>>25093
Glad to hear, yw mate

20260509_0….jpg - 165.98 KB (1340x867)

This UTAU V fork looks promising honestly. I'm not sure if it supports DiffSingers so for now I'm sticking with Lunai but wowie.

>>25293
What's the actual best OpenUTAU fork these days?

Amane_Luna.webp - 25.96 KB (250x444)

Does anybody else remember Amane Luna? She used to be relatively popular back in the day

>>25296
I think Lunai is the one everyone uses

>>25310
I do, I loved her mmd model lol. Are her voicebanks lost media?

https://youtube.com/watch?v=O1E5_-dAEbU

>>25310
All UTAUs in the Te Te Te video are enshrined in history to me. Even the pink one that sounds like a drum.

>>25333
The Nanahira one?

untitled.mp3 - 7739.85 KB

>>25310
I made a cover with her dark append, couldn't get her normal vcv to work for some reason

>>25293
Both normal OpenUtau and UtauV support Diffsinger. You just gotta download some extra stuff first to make them work

Tbh I'm not sure why people use Lunai for normal UTAUs when it was specifically made for Diffsinger use but that's a personal nitpick

>>25342
Since when? I remember you had to install a fork of OpenUTAU because the regular one didn't have Diffsinger support

>>25344
It works fine for me, but you have to download the dependencies like hifigan and stuff.

>>25342
Is DiffSinger XSY a thing in UTAUV?

>>25351
Depends on the voicebank. Most nowadays state which language they were recorded in and which ones they support through XSY.
Also again it's not just limited to V. Normal OU and Lunai also have DS XSY

>>25363
>Normal OU and Lunai also have DS XSY
I'm stupid, can you explain where to find it?

>>25376
NTA but what are you trying to find, the download for those or the XSY setting?

>>25379
XSY

ClipboardI….png - 245.34 KB (432x600)

I miss Utane Koe

Does anybody have reclists in other languages? I'm looking for Chinese, Spanish and Italian respectively.

>>26372
http:/utau.fandom.com/wiki/Reclist_Directory
Chinese, Spanish

https://youtube.com/watch?v=88Bp0yPvNyQ [Embed]

Italian (in description)

how do people change the colors of their openutau ui

>>27090
Go to preferences and change it to Light/Dark

>>27105
i was referring to the other ui elements like the piano roll. i've seen people who have their openutau in black and pink.

>>27238
recent betas of ou (incl. forks of it) give you the ability to make custom themes. so that's how, probably

>>27252
naruhodo, will download a beta then. i've been using latest lunai.

Do people still like / make Jinrikis?

https://youtube.com/watch?v=gafI6e9SuBQ

>>27428
I'm only aware of Jojo ones, but they're extremely good.

https://youtube.com/watch?v=EHBqt26YqCI

>>27428
I remember when Caseoh reacted to jinriki of himself lmao

>>27487
Funny thing is that this was probably the first instance of a Jinriki UTAU going somewhat mainstream and having original songs. Then his creator ruined his entire reputation, ruining CASE's in the process.

>>27583
If I'm not mistaken this jinriki was approved and encouraged by the "voice provider" himself. Most jinrikis never get released due to them being made without consent, but this one was. So it makes sense that a public jinriki based on a streamer would become popular.

>>27584
That being said, sucks for CaseOh that his name will forever be tied to this thing

Does anybody have DECTalk UTAU?

https://youtube.com/watch?v=apwsbYwbSho

are we gonna do it bros

>>28136
>I used Audacity and Florida Studio

>>28136
Can probably get something fairly human from a wind instrument. The human voice is just a wind instrument. Flute, recorder, saxophone, didgeridoo for example

>>28136
I think someone wanted to make a leekaloid based on printer noises a few years ago but nothing came out of it

>>28177
It's pretty quiet so I would recommend normalizing the sound samples

Efx82uzU4A….jpg - 114.56 KB (649x1200)

>>25333
>the pink one that sounds like a drum
The Kasumiga twins might be the most forgotten voicebanks of all time

>>28177
Somehow sounds more human than the other non-human loids i've heard

>>28177
Were you beat boxing into the recorder? What is that noise?

>>28180
Confirms my suspicions that wind instruments should be capable of being more human sounding. ....oh no what is going through my head. Now I am thinking you can make an UTAU from the winds of a hurricane, tsunami, typhoon or tornado. If anyone does this maybe pick one no one died in. Wait wait...the purrs and meows of a kitten

>>28181
I've seen UTAUs based on real life cats. Some unfortunately released posthumously as a tribute.

>>28182
Aw RIP kitties

>>28178
>>28180
I didn't get to hear it

What resamplers do people use nowadays? Is Moresampler still the most popular?

>>28279
Moresampler is good. Hifisampler is the fanciest but a pain to use. I’ve been using straycat-rs because Linux… I would say it’s about as good as Moresampler, maybe better, but with less flexibility/flags.

If you were to exclude Kasane Teto and Adachi Rei who would be the most popular UTAUs?

>>28509
In order of most popular to least popular:
>Defoko
>Nurse Robot
>Gekiyaku
>Kazehiki
>Poyoroid
>Ooka Miko
>All the other vipperloids
>Sekka Yufu

>>28510
TT is more popular than Defoko among producers these days so i'd put her in first place

ClipboardI….png - 867.96 KB (698x548)

Found this ancient image which lists 96Neko as an UTAULOID

Anybody know if its possible to get curve Tension working with Hifisampler?

>>28510
I'd put Momone Momo there. Barely anyone uses her but somehow she still gets a lot of fanart.

https://youtube.com/watch?v=A0cIGzlz-e8

Just found out that they're making another "clear Defoko" voicebank. There was one a couple years ago but this is a different one.

>>28286
Hifisampler in my experience is not that hard to use once you set it up. It takes a bit to render but a Diff Singer audio takes the same time.

If you're refereing to flags I've never learned how to use them so I can't comment on those.

>>30419
I don't know if it's the fact that i have a good pc but i always thought the "hifisampler takes forever to render" thing was just a meme until my friend showed me how slow it actually renders on their pc

ClipboardI….png - 32.53 KB (751x322)

So uh, do any of you have the new Lunai update? For some reason regular UTAUs are not showing up for me anymore but DiffSingers are

>>31473
Nevermind apparently they changed the "Singers" folder so now it's in Documents instead of the root folder like in previous versions

>>31473
Just installed it and it works fine. Someone really needs to tell Lunai devs to add both Riko and Rei to the Singer hub.

>>31475
i don't think they legally can add rei now that she has 3 different commercial voicebanks. with riko they could, but i dont know why they havent.

>>31476
Apparently you can request her in their dicksword but i hate joining dicksword servers so i'll pass

>>31478
If you look at the voicebank list in the singer hub it's mostly UTAU France voicebanks. So it's not as simple as joining the Lunai dicksword and requesting that they add Rei and Toriko there. In Rei's case I don't know how owns her DiffSinger, and like that other anon said I'm not sure if they're legally allowed to partner. But in TORIKO's case it's VoxSoft who would have to ask Lunai to partner with them (or viceversa).

>>31511
<how
who*

>>31511
People from the team still lurk this site so they'll probably see this. Its a matter of whether they want to partner with them or not.

>>31535
sorry, meant people from VoxSoft, not Lunai.

Someone built an accurate ust->vsqx conversion tool
https://github.com/canelumino/UtaBendr

>>31734
What makes this any different from Utaformatix?

>>31743
Its notebend based like the desc says

If vipperloid/joke utau culture were still going these days, do y'all think we'd have a Tetofan vipper like Tei was for Len fujos?

>>32090
I almost did that with my UTAU but decided against it. I really miss that era of voicebank creation, I remember Psychotic Psycho was a male yandere towards Tei and he's one of my favorites.

Any way to reduce engine noise in UTAU? Moresampler is nice but it gives all UTAUs this strange metallic sound

>>33739
If you don't mind parts taking like 10 minutes to render then hifisampler's pretty good

https://youtube.com/watch?v=V1w6KnazkDU

>>33740
Not my video and not my UTAU but here's a good comparison

>>33740
i've been using it for all of my recent hitori kanane cvvc covers

>>33739
I’ll specifically recommend hifisampler-rs on github for ease of use. Essentially, the download contains 2 executables; you run one before opening OpenUTAU and then set the other as the resampler. No python futzing necessary and works with Linux out of the box.

main.png - 513.33 KB (1920x1032)

https://x.com/matax2bi/status/2091343496145965557 [Embed]

Show the video in this Twitter/X post

"~Otomata Beta Distribution in Progress~

The automatic vowel setting and editing program Otomata has started its free beta on the 22nd.
From automatic vowel setting for Korean/Japanese VCV·CV-VC to editing, all in one program!
It's free until 8/31(Mon), so please give it a try and leave some feedback~"

http://otomata.app

>>33917
If they add ENG/Arpasing to this it would be worth money. JP functionality is cool, but honestly JP OTOing isn’t *that* bad or time consuming. Maybe it’s more for Korean though?

>>33917
>Attempt to install it
>Please insert disc 1
So we're back in the 2000's now huh

>>33917
I'm surprised no one has ever attempted to vibecode a website that does exactly this

>>33917
Why do i need to download two big extra files for this

>>34031
biruse :DD

>>34018
There should be a website where you can just create and oto an UTAU directly from the browser instead of relying on buggy software from 2008. Character files and Oto files are just text so it should be completely doable.

>>34042
I can probably do this

>>34045
You'd be a hero if you did, you know

Can the resampler from the original UTAU be used in OpenUTAU?

>>35172
If you copy resampler.exe over to OpenUTAU’s folders it should work fine

had this lying around in my computer. UTAU ver0.2.77 (from april 2011)
https://files.catbox.moe/qac418.zip

have a comparison
first one - ver0.2.77 (april 2011)
second one - ver0.4.19 (may 2024)

VB used: sakuwatari chiyo

>>35288
Am i dumb or theres no difference

>>35289
synthesis in ver0.2.77 is choppier and slightly off-tune, in ver0.4.19 it's smoother and less harsh

With so many resamplers trying to make UTAU sound realistic someone made one that does the complete opposite and attempts to make it even more robotic. It's called Chibitoy developed by infinityecho00.

https://x.com/infinityecho00/status/2098720455125540955 [Embed]

Show the video in this Twitter/X post

ClipboardI….png - 560.50 KB (1232x1164)

stupid question but how do you install hifisampler? either on base UTAU or openUTAU. i have pytorch and everything i just don't get it

>>35863
In OpenUTAU you just drop the hifisampler_torch260_cpu_win64 folder into your Resamplers (yes the entire folder) and then you run start.bat before rendering.
I have no idea if it's even possible to make it work in vanilla UTAU.

>>35807
i love it

Any clue what happened to that resampler that was based on code from V1 or V2?

>>36194
Nevermind, found it, but it seems it was never released

https://x.com/VOISTAROfficial/status/2039899236116955255 [Embed]

Show the video in this Twitter/X post

>>36195
dont wanna sound like a hater but deepvocal is a lot closer to vocaloid than this

>>36196
Well deepvocal is a whole different engine altogether, can't really compare it to a resampler

ClipboardI….png - 119.92 KB (862x813)

Is this just the wiki spreading false info or is Defoko R-18 actually banned?

>>36225
Yeah it's the wiki making up bs. The rules page they link only mentions that you can't use her for offensive or illegal works. The age also seems made up.
The Japanese have been lewding her for decades.

>>36225
First time hearing about such restrictions

HTrUnIaakA….jpg - 426.03 KB (2160x2700)

holy frick
gamechanger

>>37375
Track effects is a genuine game changer

>>37375
DAW integration seems big for original production especially

>>37375
I had fully switched to Lunai and now they release this gg

Untitled.png - 43.40 KB (556x880)

>>37386
Lunai is partly vibecoded and they also take features from the main fork anyway, sometimes before they're tested and it breaks stuff.

https://x.com/roku10shi/status/2072471107559731403 [Embed]

Show the video in this Twitter/X post

They started crediting authors since then though.
There's no reason not to switch to OpenUtau (unless you use it for lunai's own/affiliated singers and want to get updates as they come out), it's more stable and tested too. If you rely on the plugins that they have, you can just install these on OU too. OpenUtau also has custom themes now.
Another reason to switch from Lunai is because none of the OU communities can provide you support when it breaks because of the aforementioned reasons

>>37389
I'm one of those people that don't really care about vibecoding so long as it does what its intended to do and Lunai just happens to work for me but I'll check out the new version of OU. Last time I used it the DiffSinger support wasn't the greatest but maybe that has changed with this new version.

>>37375
>built-in hifisampler with curve expressions
holy shit. did they make it faster too or something?.

>>37375
BUILT IN HIFI SAMPLER OH MY GOSH

ClipboardI….png - 27.25 KB (594x431)

Getting this error while trying to install the vocoders for Toriko DS

>>37375
just tested it and holy crap diffsinger is 50 times faster to render on this version

>>38024
Are you fr? Thank god

>>38023
you can rename it into a .zip and extract and paste it into the dependencies folder

>>38032
this works. fixed.

>>37389 Streets are saying that the new version of OU is vibecoded too. Though i personally could care less.

>>38053 Why do people care so much anyway? I too use AIslop to speed up tasks like renaming 600 files or to fix my own shitty code. Besides most things today are vibecoded to some extent.

>>38053 it's not the program, it's because the infographics were made in chatgpt

>>38053 From my understanding, it's not vibecoded, but it was AI assisted on the part of one of the big devs unfortunately. I'd take AI code over them ever using AI images again but ideally the person responsible for that would get kicked from the dev team despite him being on it for years. The update has been incredibly unstable for me though so I think that's where the bulk of my frustration comes from. Perhaps I'm wrong and there was no AI code, in that case I'll be relieved