[ home / all / search / radio / downloads ] [ leek / negi / mmd / live / c ] [ g / j / meta ] [ login / status ]

/leek/ - Vocaloid Lounge

I can see this future is right now cause my voice is always going around

Name
Options
Comment
Emotes
Show Emotes


File Settings
Verification
File
Embed
Password (For file deletion.)


Thank you for coming to the 3-day meetup at Otakon 2026.


[Return] [Catalog] [Bottom]


File: 1785792674094.png(644.44 KB, 900x670, ClipboardImage.png)

 No.32179[View All]

This is a continuation of the discussion over at the Maghni AI thread but also a response to the people in the fanloid thread who were interested in the concept of "reviving" a fanloid/UTAU through a new voicebank + a redesign. Well here's your opportunity.

This is Hitori Kanane, an UTAU released in 2009. The idea is to convert her voicebank from CV to either CVVC or VCV and redesign her. Feel free to throw ideas ITT.
196 posts and 22 image replies omitted. Click reply to view.

 No.32669

>>32668
She sounds so realistic for a CVVC

 No.32670

>>32668
I guess i have a new favorite voicebank now, thanks /leek/

 No.32672

>>32653
>>32668
Awesome job both of you
We're only missing the artwork now. Are we still at the concept art stage or do wr have something yet?.

 No.32673

I think we should give her a suffix to separate her from the main voicebank (that being the 2009 one), sort of like what they did with Akita Neru NEO.

 No.32674

>>32673
Hitori Kanane RE, like IA:RE

 No.32675

>>32673
Good idea, especially if there's still plans of converting the original CV to other formats

>>32674
RE makes me think of e-mail replies. NEO makes me think of the guy from Matrix.

 No.32676

>>32675
REDUX, EVO, DX?

 No.32677

>>32676
i like Hitori Kanane REDUX the most personally

 No.32678

>>32675
More of a word, but Perennial is used to mean a plant survives over winter, which applies to periwinkle flowers

 No.32679

>>32678
Feels a bit long, maybe it could be shortened?

 No.32680

>>32672
Still at concept stage but we sort of have an idea of what the final design should look like
We only need someone to draw a basic sketch of some of the ideas ITT and i'm guessing Sora will do the final artworK(?)

 No.32683

Tested the new voicebank (without the merged audios from >>32668 because i have no idea how to do that) with hifisampler and it sounds absolutely amazing

 No.32690

I meant someone should put the 2 pitches together lol

 No.32692


 No.32693

is it me or she's missing the ガ

 No.32697

>>32693
She's encoded in hiragana, so it should be が

 No.32698

First to make a cover with her

 No.32699

>>32676
Voting for Hitori Kanane REDUX

 No.32700

>>32698
God she sounds so amazing. Also is it me or you made her slightly higher pitched?.

 No.32701

>>32700
yeah i tried to make her sound closer to the original

 No.32702

>>32699
REDUX feels too long. I vote for RE.

 No.32703

I know its "Evolution" but EVO makes me think of a robot from Sonic but its the best option here. Hitori Kanane EVO / 悲音ヒトリEVO.

 No.32704


 No.32705

File: 1786245920509.png(92.48 KB, 394x146, ClipboardImage.png)

I don't know if the person that made Ena's logo is still here but if you see this post could you remake Hitori's in a similar style?

 No.32706

>>32607 >>32629

Haha, yay! I was thinking earlier in the thread that I could probably get pretty close and I remember the exact tone I used when recording her (I feel like it's almost become a stim at this point, lol), but I figured it'd be kind of silly to say I thought I sounded similar <15 years later. Happy it sounds close to you guys too!

>>32657
I wasn't sure when recording if I should prioritize smoothness or enunciation or not- if there are any issues, I can record more takes, just lmk :) The n-n string was particularly mushy so I took that as a cue to redo all the vowels and make them clearer haha

I feel like my t's are a bit strong too?? In any case, I'm not opposed to recording more if there's feedback.

>>32668

What a treat to wake up to! Thank you so much! Haha I was listening to the old kenkawakare included in her og vb while recording and got emotional, how nice to have an update

>>32698
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA

>>32678
I think this is so cute but there's no way to shorten it a bit, huh :<. Flowers..!

My two cents on the differentiators is that I am a bit biased towards :re as her original unit was r2 (yes.. like the geass season...) so two letters with r looks pretty natural to me and like an evolution of that!

 No.32707

>>32698
She sounds similar to Nurse Robot Type T

 No.32708

>>32706
EVO seems to be winning the poll

 No.32709

>>32708
Hehe I'm fine with deferring to whatever wins ^^

 No.32710

>>32706
Glad you liked the cover, because i'm going to be making a lot more with her. I mainly do covers with TORIKO, a voicebank that was made on this site for April Fools, but I'm really liking Hitori and her new voicebank.

 No.32712

I love her she's so high quality

 No.32713

>>32712
this song makes me cry
also wow she does sound really high quality, did you do some processing?

 No.32714

>>32713
I used Hifisampler to render so that might be why

 No.32715

OTOanons, can we get a new version of the CV >>32183 with OpenUTAU compatibility (character.yaml) and a less kusopy oto?

 No.32716

>>32715
<kusopy
meant sl0ppy. i forgot we have a wordfilter that changes sl0p to kuso.

 No.32717

I know EVO is the main focus right now but does the possibility of an English UTAU voicebank exist?

 No.32718

>>32710
Awww thank you so much! <333 I'll be looking forward to it!!!

>>32712
AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA

---

If I can make a request could anyone do ((even a snippet) of palette or goodbye days 👉👈))

 No.32719

>>32717
I heard making english UTAUs is a pain so most people go the DiffSinger route. But since DiffSinger is a type of AI voicebank I'm not entirely sure if the voice provider would be fine with it.

But either way, like you said, the current focus is EVO so let's work on that first then if Sora's down to record more Hitori voicebanks we can try something.

 No.32720

>>32719 >>32717
I'd have to look into the reclist- I was worried about how long it would take me (because I remember 1.0 being an odious, long affair...) so I just did the standard cvvc but it's painless enough now that I was able to do two pitches relatively quickly and I'm kicking myself a little bit for not just committing to the expanded list from the beginning, haha.

I've seen diffsinger brought up a few times so I'll do some research into it before I commit to an answer.

My general stance for the UTAU vb though is that I'm open to recording more if there are specific requests for things!
(I vanquished the hard R sounds that haunted me in 1.0 only to be pursued by t's in 2.0 huh,,,,,)

 No.32721

>>32720
DiffSinger in few words is basically the lovechild of SynthV and UTAU. If you can imagine SynthV but with the ability to create your own voicebank it's essentially that. Most voicebanks are ethical because they're created with the voice provider's consent, and since the AI is trained and used locally you're not using millions of liters of water to power some random datacenter in Memphis.
https://vocadb.net/T/9389/diffsinger
Here's a sample of what Toriko Diffsinger sounds. She was made on this site:
https://www.youtube.com/watch?v=fbAoOg-back
>(I vanquished the hard R sounds that haunted me in 1.0 only to be pursued by t's in 2.0 huh,,,,,)
The T's sound fine to me

 No.32722

>>32207
I am so glad that one anon told you to email the VP.
Inactive socials mean nothing.

 No.32723

>>32722
Yep, and now we have Kanane Hitori EVO

 No.32725

>>32721
Thanks for the info! I'll think on it :)

(tyty Maybe it's just a me thing then! I also thought they sounded fine in the raw recording but regardless of what I'm doing I always nitpick myself anyway haha. If they sound fine to others I'll kick the thought from my head.)

 No.32726

Thank you for everything Sora
I really like Hitori and her EVO voicebank

 No.32727

>>32726
Thank *you* for liking her! She's alive again because of everyone's love <3 What a special revival for a girl named for being alone.

 No.32728

>>32727
You're welcome anon
Sorry that it took 17 years and the creation of a bootleg 4chan for Hitori to get this much love

 No.32729

>>32727
How does it feel to be in your 30's and revisiting a project you made early into your 10's only for it to get more attention than when it originally came out?.

 No.32730

>>32722
you'll never guess who said to do that.................

 No.32731

>>32729
Honestly, even back then every instance of her being used felt like a gift and not much has changed in that aspect. I’m mostly just thrilled to be getting content for her even after so many years!

I’ve always engaged with UTAUs because I love vocal synths and the collaborative creativity that comes with their music (songs, storylines, MVs, etc), so I’ve never had any goals or metrics in mind for how much she gets used or how popular she is. It’s a happy surprise each time. If I think back to 2009 I was almost using her as an avatar in a group space, so I don’t think I ever had any expectations for people to pick her up or for her to trend (aside from thinking in my heart like wouldn’t that be cool, haha)

Hitori’s gift to me has always been community, now and in the past.

 No.32732

>>32731
>collaborative creativity
You're in the right place
https://utau.fandom.com/wiki/LEEKAloid

 No.32733

>>32731
If you ever make an English voicebank whether its a regular UTAU or DiffSinger I promise I'll give her an original.



[Return] [Catalog] [Top][Post a Reply]

Delete Post [ ]
[ home / all / search / radio / downloads ] [ leek / negi / mmd / live / c ] [ g / j / meta ] [ login / status ]