Mira a twich streamer

[timber thistle]

PLEASE IM FUCKING BEGGING YOU

[sage widget]

💡 Tip: You can try using the /search or /find command from @trail pier or @empty otter to check if someone already made this model. Alternatively, you can check the #1175430844685484042 channel or use weights.gg, but keep in mind that weights receive the models after us, so if something new comes out, you'll find it on our server first.

[timber thistle]

I can send a yt vid on her voice

please thi

https://www.youtube.com/watch?v=DLvELwNHCHs

pliz guys i beg you

ill suck ur toe nails

[fallen oak]

If you can prepare the dataset ill train it for you

[timber thistle]

dude

can you prepare the data set pls i dont know how i can do it

[timber thistle]

i can train it

ill watch some yt vids or smth

pleaseman

i just dont understand how i can get the voice of her in the data set

[fallen oak]

do you have a good gpu? cuz training without a gpu or online on google colab is hell of a trouble

[timber thistle]

just tell me how to make a dataset

ill figure out the rest

😦

[timber thistle]

also nah idh a good one but i think its decent

[fallen oak]

you can find tutorials online, you basicly just have to seperate vocals from instruments get the raw vocals and remove any breathing sound ( or any unnormal sound ) using an audio editor like audacity

[timber thistle]

THATS THE FUCKING THING I CAN DO ALL THAT BUT I DONT KNOW HKOW TO GET THE VOICE OF HER

GAH

https://www.youtube.com/shorts/Auzt875U0pI theres this vid as well

[timber thistle]

do this voice please there r many other vids of this specific voice on that guy's channel

https://www.youtube.com/shorts/E0D53JtdRxY

DO THIS

PLEASE

IM

FUCKING BEGGING YOU

IVE ENVER BEGGED THIS MUCH IN MY ENIRE LIFE

ENTIRE

[timber thistle]

@fallen oak

[fallen oak]

bro alright alright were gonna make the model theres no rush

[fallen oak]

what do you mean?

[timber thistle]

Yayyyy

[cloud forum]

Ayo? @timber thistle level 5 !!!

[timber thistle]

!!!!!

thank you

so fucking much

[timber thistle]

if you can, please do this one

[fallen oak]

well i can train it for ya, but can you prepare the dataset?

[timber thistle]

well

uh

can you tell me how i can get a recording of her voice

[fallen oak]

why are you asking me that? just find vids where she speaks in, download the audio from the vid, then cut out any other voice ( other than her voice )

then just run it through an AI online to seperate vocals from instruments

[timber thistle]

How the hell can i download the audio

[feral spire]

I use 4k video downloader, but the pros use ytdlp. Another simple solution is just remove "ube" from the youtube link and you will be redirected to a downloader page.

After you download, you need to remove any unnecessary audio from the video, like background music or other voices. To remove people talking, I recommend vocalremover dot org, but you might need to do it in batches. It's probably better to use UVR5 but idk how to use that. To remove other voices you'd need audacity or something.

[timber thistle]

oh..

aight

[feral spire]

Wait is she cottontailva?

Like the one from r34?

[timber thistle]

uh

what?

[feral spire]

@timber thistle

[cloud forum]

Ayo? @feral spire level 1 !!!

[timber thistle]

like

shes a

h*ntai voice actor

;-;

im tryna

[feral spire]

Well she introduced herself as cottontailva

[timber thistle]

yeah ik

how do you know

[feral spire]

I'm a fan of her Judy Hopps voice

[timber thistle]

oh

rlly?

i just found her randomly

Lol

anyways

[timber thistle]

thanks for this tho

[feral spire]

lol i thought you were a bigger fan

[timber thistle]

dang i dint know she voice acted judy hopps

[timber thistle]

oh nah i just searched mommy voice and this came up

[feral spire]

On rule34 she did

not the actual movie lol

tho she is talented enough for that too

[timber thistle]

yup

it downloaded!

lemme look up a vid on how i can use it

bhahahaha

[feral spire]

ah ok. Well you're gonna need a lot more data to make a good voice model.

[timber thistle]

rlly?

[feral spire]

Maybe 20 mins would be good

[timber thistle]

bruhhh

Oh bruh

how can i get

that

is 7 mins good?

gosh idk

[feral spire]

im looking around as well

[timber thistle]

@feral spire

[timber thistle]

thank you man

for your help

https://www.youtube.com/watch?v=2Kvi-yEA-48 is this her?

[feral spire]

yup I think so

[timber thistle]

les go!!

its gonna be tough to remove other vocals tho

[feral spire]

but she doesn't seem to do the mommy voice you wanted. So the end result will sound more or less like the normal version of her

[timber thistle]

yeah..

im so fucked

god dang

what do i doo!1

1 min wont be good?

[feral spire]

There are quite a few mommy voice models though

[timber thistle]

yo

i seperated the instruments from that video

but how do i seperate other vocals

[feral spire]

download audacity and clip them out

[cloud forum]

Ayo? @feral spire level 2 !!!

[feral spire]

https://www.weights.gg/download?fileUrl=https://models.weights.gg/clqu3hfjf004vs4jv4177kx8c.zip

this is one but i haven't tried it yet

[timber thistle]

o thanks

i made a

file

idk if it will be good enough tho

can you help me with audacity?

i can send you the voices

[feral spire]

There's already a cottontailva model here, I should have checked earlier lol
https://discord.com/channels/1159260121998827560/1202212677376802860

[digital heart]

ayo that conversation is wild ;-;

[timber thistle]

thats kinda sh

it

can u just help me with audacity

please?

[timber thistle]

3cf

[cloud forum]

Ayo? @timber thistle level 6 !!!

[timber thistle]

@feral spire

[timber thistle]

@feral spire

[feral spire]

Kinda working on another project right now, are you sure the model isn't good? I'll check it out

Hmm the data set was from her stream apparently, and I guess we need some more "diverse" data

[timber thistle]

dang man\

this shit real confusing

https://www.weights.gg/models/cls1pa86n004h5qnhdcswab6p i found this one

i think its pretty accurate

[timber thistle]

@feral spire