#I need help
1 messages · Page 1 of 1 (latest)
What is your PC GPU? There's a newer W-Okada version.
5070 ti
im on the nw update
Try Tg Develop's W-Okada fork. https://docs.aihub.gg/realtime-voice-changer/local/tg-develops-w-okada-fork/ https://github.com/tg-develop/voice-changer/releases/tag/b2397
I thought
I was using this yesterday and it just stopped today giving me that notification
Older W-Okada versions like v.1.5.3.18a and b2332 are not known to work with GeForce RTX 50 series GPU.
so this verision will work perfctly for my gpu?
b2397 W-Okada will work with RTX 50 GPU.
which one do I download for nividia gpu
Download voice-changer-windows-amd64-cuda.zip.001 and voice-changer-windows-amd64-cuda.zip.002.
question why both
and do I need both?
These files are split zips of a single zip, these two files have to be in the same folder, and use WinRAR or 7-Zip to open the .zip.001 one. I've been saying this for many times, though I doubt if people ever remember this step.
can u help me set it up?

lol
Until you open MMVCServerSIO.exe.
That's another part the split zip of a single zip file. If you open .002 (.003, .004 .... .00x) with WinRAR or 7-Zip it would throw up error.
so just only use the 001
and keep 002 untouched
?
@lofty wren also how do i run this
Check one of your browser, see if there any 127.0.0.1:18888/.
ther is
but it stopped
how do i add models
@lofty wren
@lofty wren I wennt back to ther old one but why is it not working

what
The nw one way way to cnfusing for me
Who told you to use the older version?
and the old one was working for me yesterday
No1
but why is it not wooorkling ive ben using for a whol 3 months
This is Tg Develop's W-Okada fork. https://cdn.discordapp.com/attachments/1159290139609137264/1457950706500763809/image.png?ex=696d0727&is=696bb5a7&hm=5a444eb37f24bc06dfb8599c5583106c0bb591583b0fe922d4451363f107368a& If you can't open the program, review your steps again.
yes I understand but for me the older one is not complicated for me and the localhost is confusing
Please stop using the older version, now.
I even open this W-Okada version for over 69 times, and I still found no problem. If you were to extract the zip file in "Downloads" folder, try extract to somewhere like "D:\MMVCServerSIO".
I got it
but
where can I put my models
it doesnt save them
This is where you upload RVC voice model on Tg Develop's W-Okada fork.
ok ty
its not coming out of my
Your settings:
Chunk: 128 ms
Extra: 2.7 s
GPU: NVIDIA GeForce RTX 5070 Ti
Pitch extraction (F0): rmvpe
Input: microphone
Output: Line 1 (Virtual Audio Cable)
Monitor: your speakers/headphones
CAble input
If you're wondering why there's no "Line 1 (Virtual Audio Cable)" in your system but I say like that, that one is from a different program that works similar to VB-Cable but still.
i fixed it
ty
this so much bettr
its on a localhost now
and when i screenshare thers no AI logo now
W
Is this solved?
yea but my sets from old thing are WAYYY to deep
@lofty wren
how can I change it to harvest
@lofty wren How can I put this into my shit
I have it in my sets
and whers harvst
cxause it sounds completeeely differnt
Why you still using this version?
rmvpe is generally better than older F0 models like harvest and crepe for most scenarios. But when you go in "advanced settings" then "Downloader" and scroll down there, there's no "harvest" anymore.
IM not
i want to convert
it was a video
Any index ratio of any non-zero value will strain CPU performance. It's a quirk in every W-Okada version, even in v.1.5.3.18a, not just W-Okada forks.
Even in b2332 W-Okada fork, there's still no F0 pretrain model "harvest" there.
damnnn