#CPU TO GPU
1 messages · Page 1 of 1 (latest)
did you download the right program
yes
which one
MMVCServerSIO_win_onnxgpu-cuda_v.1.5.3.18a.zip
what kind of graphics card do you have
GPU:amd radeon RX 5500 XT
CPU: AMD Ryzen 3 3100 4-core processor.
Ahh, where do i get it?
same place where you downloaded the onnxgpu_cuda version
ye
Okay, i'm gonna attempt
make sure to uninstall the other version first
ye
okay, thats what i did
yup
thank you so much!!
ofc
That is over a year old lol
It's a really old version of original wokada
Using the latest Wokada Deiteris fork will give you better performance, may want to try that on a new help post or want the link there?
Basically, Wokada has 2 main versions
- original by Wok
- deiteris fork (modified version) made by deiteris
The latest deiteris' fork is suggested for best performance and also to use fp32 mode which helps with quality
-realtime
Guides for Programs that use RVC Models in Realtime for Calls/Games
Most suggested. GUIDE
ONLY the latest alpha comes close to the Deiteris Fork performance, older versions in youtube tuts are way worse. GUIDE
Unavailable, the guide is outdated and the program is worse compared to the ones above, and much less updated
1st link
If you have vb audio cable too, uninstall it, many users reported issues with it
That old wokada version along with vb audio cable are common things said in video tutorials, which are pretty outdated
Yup
okay its extracting atm
also
the vb part
i'll look into it, bc like VB is like my only good source of audio lol
unless you have something i could use?
for audio
Vac lite at the 3rd step of the wokada deiteris fork is better for not running into future issues
Yes
I'm suggesting you this since other users on windows reporting vb audio cable to just stopping to work
It's not a website, it uses a Web User interface, check https://rentry.co/forkvoicechangerguide#why-does-it-run-in-a-browser-and-not-its-own-window
It's a web user interface, that's the thing, please check what I sent you
ok
Because it uses a Web User Interface (WebUI) coded in JavaScript & TypeScript, the majority of AI programs are designed to run on the browser since it can be used both on cloud and locally (on your pc), and it can also resolve some performance issues. The original wokada also ran on a WebUI, just that it made it's own window
TL;DR: the user interface is coded on the web bc it's easy to update + works on cloud + fixes performance issues
Mhm, alr
Of course, it still all runs locally on your hardware, just the interface is a web ui
Run setup64.exe
mhm
Then when it asks to open the audio device settings, click yes, and set your recording and audio usual devices as default ones
uh, i'm so lost now
You need to select your usual audio device, then click as default
I mean I can't know what you usually use for example outputting audio, could the that 1st element that seems like a TV in your screenshot but that depends on what you use personally
okay, i did
mhm
Do this for the recording tab too
Great
Just click ok, go to the windows app settings and uninstall vb audio cable while you're at it
Just if u want to free up storage
so VB audio cable is not being used anymore?
Yup
No one really suggests it anymore
Bc most staffers saw that users keep reporting the same issues with it on windows
downloaded vac lite already
Yeah, I was just telling you that the other 2 things you mentioned weren't needed anymore
i don't think vac lite pops up in the settings
alr
Close Wokada deiteris fork, reopen it, be sure that you gave it microphone permissions, then in output you should see a line 1 option
If you do
Send a screenshot of your whole wokada so I can suggest the rest of the settings
Might need to not have leftovers of that program or issues
alrighty brb then!
Cya
i just deleted it
and its still here
i have to delete both
just noticed that lol
okay
restarted againb
i'm back
what do i do now?
Open wokada
Show a screenshot of ur wokada so I can suggest u settings
Chunk: 192ms
Extra: 2.7
Input: microphone
Output: line 1
Monitor (to hear urself, optional): headphones
Also you could try force fp32 mode in advanced settings which makes models more stable and better in quality but gives a bit more delay, your choice
Click edit, upload the model files
okay did, but was does it sound like a robot and why does it pick up other app audios, its picking up my friends mic on discord
Enable sup2 and echo
Also, you might wanna try other models too, not all models are high quality it depends how they were trained
You can also try this too for better quality
Did you upload a model, selected it, and click start?
What browser are you using?
Opera GX
The output has to be line 1
If it still has issues, I might suggest trying other browsers like Chrome or Firefox to see if they give the same issue
Line 1 (Virtual Audio Cable)
nope!
Alright, have a nice day vox lol
Thus will be marked as solved, for future issues pls make a new pist