im not sure. I downloaded the Homer Simpson Model, and it sounds NOTHING at all like what was shown, so I just have no idea what's wrong, I thought maybe I needed a special embeder or something, but I guess that isnt it
#✨│ai-help
hallow thistle
im not sure. I downloaded the Homer Simpson Model, and it sounds NOTHING at all ...
Well, it sounds like you didn't set anything right on GUI.
eager elk
I set pitch and index ratio, selected the correct GPU, set up everything to use WASAPI, selected RMVPE_onnx wtc. Not sure what else there is to set upp
hallow thistle
You might too expect for best known working settings, which is what I'm going to tell you:
Chunk: around 70 - 110 ms
Extra: 2.7 s
GPU: AMD Radeon RX 6950 XT
Pitch detection (F0): rmvpe_onnx
Input: your microphone
Output: Line 1 (Virtual Audio Cable)
Monitor: optional, you can route this to your speakers/headphones to hear the program.
An index file is a file that stores accent of a model, typically comes alongside pth file which is the actual voice model file. It is commonly used in Retrieval-based Voice Conversion (RVC) program (like Applio RVC fork). In W-Okada, however, the "Index Ratio" is supposed to be value 0, as index is not needed in voice changer; by setting it to any value, the program would still work but also use more GPU resources to process.
eager elk
An index file is a file that stores accent of a model, typically comes alongside...
ah, so the index file won't actually add an accent to the voice in W-Okada? That makes more sense then.
tough fiber
anyone has vbcable audio driver file. their own site not reachable atm
-vbcable
<@&1159293204038955078>
low shard
anyone has vbcable audio driver file. their own site not reachable atm
we do not suggest vb audio cable for realtime voice changers on windows
what's your exact pc gpu and os?
dreamy seal
It's been awhile since I used rvc,which one still works with rvmpe?
hasty jay
we do not suggest vb audio cable for realtime voice changers on windows
Can you help me aswell?
ashen plank
Hey guys, I'm trying to set up Applio for my AMD GPU so i can train voice models locally, the issue is that it's not showing my GPU : Unfortunately, there is no compatible GPU available to support your training. And when I run it it says : An error occurred extracting the index: need at least one array to concatenate
If you are running this code in a virtual environment, make sure you have enough GPU available to generate the Index file.
GPU : AMD Radeon RX 6900 XT
System : windown 11
and I'm following this totorial : https://docs.aihub.gg/rvc/local/applio/#amd-on-windows-precompiled-fix
viral mason
It's been awhile since I used rvc,which one still works with rvmpe?
They all do, use refinegan it's the best new pretrain and the only one recommended to be used other than the original
Refinegan is spinv2 and is only available on the newest branch of applio
hasty jay
I downloaded VAC, i can hear but it wont register my voice. what do i do?
peak pebble
GPU: 3070ti Windows 11. Which version of okada should I download and from where?
viral mason
GPU: 3070ti Windows 11. Which version of okada should I download and from where?
u should use vonovox
cerulean pond
Who can send me all their RVC files, please? (via gofile)
tame oracle
!howtoask
patent trellis
sweet hedge
yo i got rx 5500 xt
r5 5600
windows 11
16gb ram
which one should i download for good voice?
peak pebble
u should use vonovox
Thank you
GPU 3070 TI Windows 11 I tried to download Gawr Gura voice model but it keeps showing invalid password and username. Why is that?
viral mason
GPU 3070 TI Windows 11 I tried to download Gawr Gura voice model but it keeps sh...
what
wherever you're getting the voice from should not have a password or anything
definetly a scam site
get all voices from here ^^
https://discord.com/channels/1159260121998827560/1175430844685484042
peak pebble
Oh okay! Thank you
viral mason
btw here is the guide for vonovox
https://docs.aihub.gg/realtime-voice-changer/local/vonovox/
peak pebble
btw here is the guide for vonovox
https://docs.aihub.gg/realtime-voice-changer/l...
I will go through it. Thanks
knotty moth
what's your exact pc gpu and os?
he has been asking the same crap since before, and he said it rtx 50-series
viral mason
bro does not need deiteris if he has a 50 series
tame oracle
bro does not need deiteris if he has a 50 series
he either means a 50 series card or literally 5000 GPUs
fallow python
young halo
uh, what am i doing wrong
hallow thistle
he has been asking the same crap since before, and he said it rtx 50-series
Bro either still can't get over it, not reviewing his queries or having a dementia.
hallow thistle
yo i got rx 5500 xt
r5 5600
windows 11
16gb ram
which one should i download for...
See Tg Develop W-Okada fork. https://docs.aihub.gg/realtime-voice-changer/local/tg-develops-w-okada-fork/ https://github.com/tg-develop/voice-changer/releases/latest
young halo
uh, what am i doing wrong
help :c
hallow thistle
Who can send me all their RVC files, please? (via gofile)
For voice models, there's #1175430844685484042. If Hugging Face is blocked in your internet from your country, consider VPN or check again. If you mean something else, please be specific.
hallow thistle
I downloaded VAC, i can hear but it wont register my voice. what do i do?
If you can't "hear" yourself on W-Okada, the issue won't be the Virtual Audio Cable software itself. Rather, there's something to do with audio settings in the voice changer. https://cdn.discordapp.com/attachments/1159290139609137264/1446171264489357322/image.png?ex=69330372&is=6931b1f2&hm=756cdc25641223097fca41fe2b520109fc818e740befb445d10c8e5e56c83e06&
ruby steppe
I need help with the voice
i turn the voice changer on, my mic is set on the right stuff. but i cant hear it
i have vb audio
ruby steppe
@viral mason how to get vonovox audio to discord
i cant upload images
I can hear myself in the voice changer
hallow thistle
<@1023278814752677918> how to get vonovox audio to discord
Use helper role instead of one specific user.
!howtoask
ruby steppe
Use helper role instead of one specific user.
oh well can u help me
<@&1159293204038955078> . I set my input on discord to. Line 1. And I cant hear anything at all
hallow thistle
You would also read help guidelines before start asking anything.
hallow thistle
Don't be too hurry. I'm trying to provide an answer.
ruby steppe
Don't be too hurry. I'm trying to provide an answer.
ok mb
hallow thistle
In Vonovox, there are only input and output. You can set "output" to your speakers/headphones to hear the program, although this won't output any audio into Line 1 at the time.
Virtual Audio Cable and VB-Cable are two different programs. I don't use VB-Cable, but in my screenshot it's Virtual Audio Cable.
ruby steppe
my vonovox dosnt look like urs
hallow thistle
Either because mine uses light mode or the version being v1.6.9. Check again.
ruby steppe
check dms
hallow thistle
You can send an image on here now. 
ruby steppe
hallow thistle
By the way, Vonovox and "W-Okada" are two different programs, but both function similar to each other (being realtime voice changer). Don't always confuse between these.
This is the actual link to Vonovox. https://github.com/dr87/Vonovox
ruby steppe
This is the actual link to Vonovox. https://github.com/dr87/Vonovox
so i download this right
hallow thistle
Don't be too hurry. The "W-Okada" in your screenshot is an outdated one. Vonovox in question is more preferred, although there's another newer W-Okada fork version that's more UI friendly than Vonovox.
ruby steppe
Which one should i get then
hallow thistle
Vonovox.
ruby steppe
ok im running the setup to download the requirements
ruby steppe
ok im running the setup to download the requirements
almost 10 mins and its still downloading the requirements dang.
hallow thistle
If your internet is slow, it's kind of expected.
ruby steppe
good settings?
Nevermind, i just need this to work in discord and wtv game I want
hallow thistle
good settings?
Here's your Vonovox settings:
Block size: 0.10
Extra tme: 2.00
Crossfade duration: 0.15
quasi gyro
how can I enable fp16 on an old fork of applio, need it for an experiement
pulsar pasture
hey would anyone be able to help with this?
ah dam can't post pictures yet
basically i cant seem to select my gpu, it just has options like cpu, gpu0, gpu1, gpu2, ect
it doesnt have a drop down menu like ive seen in the videos
im using a amd 9070xt btw on windows
viral mason
basically i cant seem to select my gpu, it just has options like cpu, gpu0, gpu1...
I smell an old voice changer
pulsar pasture
ah, i did just notice the program name is actually title "voice changer client demo"
so that might have something to do with it lol
viral mason
Since u have amd I'd say download wokada tg fork ^^
It's that second guide the one right under Vonovox
-rt
viral mason
There are multiple versions on the download page for it and you'll also need vac lite (it is like vb cable)
I'm heading to sleep so if ya need more help ask a mod or helper, I'll be able to help when I wake up tho
pulsar pasture
real quick whats a vb cable?
pulsar pasture
i havent taken a look at it yet sorry
viral mason
It's alright, you only need vac lite plus the voice changer anyway
pulsar pasture
i assume by your response its explained in the link u gave
pulsar pasture
chillin, thanks gamer
viral mason
And if u get confused that's why there are mods and helpers :3
viral mason
chillin, thanks gamer
Np man, gn
viral mason
Not yet lol
shell crescent
When weights will bring back text to speech?
hushed steppe
marsh nest
can someone help me with the voice changer ?
i did everything that i saw in youtube but still its not working
like i hear sudden unrecognised sounnd so im asking for help
sweet hedge
how to find bestq pitch and formant shift settings?
because everything sound terrible to me
obtuse summit
so i am on ubuntu 24.04 can anyone help me setting it up?
deiteris W okada fork
local
can anyone help me?
plz
i need help
PLZ HELP MEE
DUDE IT DOESN'T WORK
omg bro
this program is broken
only windows user can have stuff like this
...
primal valley
so i am on ubuntu 24.04 can anyone help me setting it up?
deiteris W okada fork ...
@obtuse summit Did you make sure to install the linux version?
obtuse summit
i am on linux rn..
and yes i installed it
Gpu nvidia and cpu won't work
i can't use the second cause i don't have another gpu
i have an nividia gpu
primal valley
both files installed?
primal valley
aight, I got nothing then
Idk linux 
oh, and to answer your other question in chat
Models are usually trained on their default language, so they can do other languages
but some sounds unique to certain languages will cause clashes when speaking
there do be some models that are trained in italian tho, but you can also try english ones and see if they work for you
@obtuse summit https://discord.com/channels/1159260121998827560/1364589802921660527
Check this one out
obtuse summit
oh, and to answer your other question in chat
Models are usually trained on the...
thx i am trying it now
txh
it worked
so
it works but i can't gte the model to work
nv imma give up
simple helm
I downloaded(MMVCServerSIO_win_onnxgpu-cuda_v.1.5.3.18a) but there is no RVC v2 where to download it
viral mason
When weights will bring back text to speech?
ask in the official server
viral mason
where i download the voice changer?
what is ur gpu and os (windows, linux ect)
viral mason
I downloaded(MMVCServerSIO_win_onnxgpu-cuda_v.1.5.3.18a) but there is no RVC v2 ...
that's super duper old, what gpu do u have and what is your os
torpid nacelle
what is ur gpu and os (windows, linux ect)
RX 580 2048sp, win11
viral mason
RX 580 2048sp, win11
since u have an amd card I'd say use wokada tg fork
the guide is the second one , it's right under vonovox
-rt
tame oracle
!howtoask
torpid nacelle
since u have an amd card I'd say use wokada tg fork
fine
warm marlin
Hi!
My GPU is : RTX 5070
My operating system is windows 11
I canno't hear anything from the voicechanger, seems that it doesn't work
Tutorial Used: https://www.youtube.com/watch?v=SxdnGxicJOg&t=124s
If somebody can help me 🙂
viral mason
you have
an outdated software :D
since ur gpu is super good I'd say use vonovox as it's the best voice changer currently for nvidia
it's the first guide
-rt
viral mason
and u will need these two downloads to use it
https://github.com/dr87/Vonovox/archive/refs/tags/v1.6.9.zip
@warm marlin
warm marlin
so it's not the same software ?
viral mason
so it's not the same software ?
it is, it's better and new
like way better
it's a realtime ai voice changer but not old and outdated
atomic solstice
my voice is pretty laggy on an amd rx 7700xt and im on windows 11
warm marlin
it is, it's better and new
and do you have any youtube video to recommand me ?
like a tutorial
tame oracle
!howtoask
atomic solstice
my voice is very laggy
rx 7700 xt
Windows 11
tame oracle
my voice is very laggy
rx 7700 xt
Windows 11
please read the help guidelines above before asking
viral mason
and do you have any youtube video to recommand me ?
No all YouTube tutorials are outdated
quasi condor
so uh what's your PC GPU?@twin spruce
quasi condor
is it gonna mine bitcoin
uh no what 😭
quasi condor
i suggest u don't use the voice changer while gaming on high intensive games
twin spruce
i want to sound like sponge boob
twin spruce
i suggest u don't use the voice changer while gaming on high intensive games
does roblox count?
quasi condor
does roblox count?
roblox is fine as long the graphics quality is low or medium
twin spruce
ok
twin spruce
but what the app for the voice changer is
twin spruce
ok
quasi condor
ok
also you can find the spongebob models on https://discord.com/channels/1159260121998827560/1175430844685484042
twin spruce
thanks :3
viral mason
https://docs.aihub.gg/realtime-voice-changer/local/deiteris-w-okada-fork/
that one isn't really recommended as much, I'd suggest wokada tg fork
I switched from deiteris to tg fork and tg fork I say is better
quasi condor
that one isn't really recommended as much, I'd suggest wokada tg fork
the hell is that o- o
viral mason
-rt
viral mason
second one :p
quasi condor
second one :p
eugh spin embedder
viral mason
it supports content vec models goofy
doesn't need spin
switching models on it is almost instant
quasi condor
it supports content vec models goofy
dude the girl has a 1650 do u expect its gonna handle the modified version of okada?
viral mason
dude the girl has a 1650 do u expect its gonna handle the modified version of ok...
I didn't know what gpu they had my bad 😭
viral mason
What version should i chose
stick with deiteris, the one balbal first gave u
twin spruce
Deiteris fork?
viral mason
ye
quasi condor
I didn't know what gpu they had my bad 😭
that's why u gotta read the message first 😔
lmao
twin spruce
voice-changer-windows-amd64-cuda.zip.001
1.95 GB
Dec 7, 2024
voice-changer-windows-amd64-cuda.zip.002
613 MB
Dec 7, 2024
voice-changer-windows-amd64-dml.zip
What is the difference between these 3? @quasi condor
simple helm
that's super duper old, what gpu do u have and what is your os
I have an RTX 4060 and a win 11
simple helm
where and what to download?
twin spruce
theres too much steps for my brain to comprehend
simple helm
for RVC v2
viral mason
I have an RTX 4060 and a win 11
u should use Vonovox since ur gpu is good enough to handle it ^^
https://github.com/dr87/Vonovox/archive/refs/tags/v1.6.9.zip
https://software.muzychenko.net/freeware/vac470lite.zip
the guide is here in case u need help and u can ask me too
https://docs.aihub.gg/realtime-voice-changer/local/vonovox/
for vac lite, run setup64
simple helm
is it better than the one I downloaded? (for a female voice)
viral mason
is it better than the one I downloaded? (for a female voice)
it's currently the best of the three up to date ones
if u got one off yt it's old and outdated by over a year
simple helm
does it change in real time too?
simple helm
u should use Vonovox since ur gpu is good enough to handle it ^^
https://github....
so I just download it and it already works?
viral mason
did u read the guide at all?
quasi condor
voice-changer-windows-amd64-cuda.zip.001
1.95 GB
Dec 7, 2024
voice-changer-windo...
sorry uh im cleaning the house one sec
viral mason
so I just download it and it already works?
for vonovox run setup, then run start
for vac lite run setupx64
quasi condor
voice-changer-windows-amd64-cuda.zip.001
1.95 GB
Dec 7, 2024
voice-changer-windo...
click this one
twin spruce
ok
simple helm
I'm sorry, I just don't understand English well.
quasi condor
I'm sorry, I just don't understand English well.
what language do u speak
simple helm
Rus
quasi condor
I'm sorry, I just don't understand English well.
он сказал, вы читали руководство?
twin spruce
have u also installed the virtual cable?
no
viral mason
он сказал, вы читали руководство?
translate this to russian for them
for vac lite run setupx64
for vonovox run setup, then run start
quasi condor
translate this to russian for them
for vac lite run setupx64
for vonovox run set...
aight
viral mason
thx
twin spruce
it said that i had to use a command on the terminal too
viral mason
here's vac lite btw https://software.muzychenko.net/freeware/vac470lite.zip
quasi condor
@simple helm
Он сказал, что для Vac Lite запустите Setup x64.
Для Vonovox запустите Setup, затем запустите Start.
simple helm
Show me what the interface should look like, otherwise I don't understand it well.
it writes to me when I press start (Error
Cannot start voice conversion - No
GPU available. You can still
access settings and configuration.) but I can't find where to put the graphics card.
ivory juniper
ffmpg not found how to fix?
vale canyon
ffmpg not found how to fix?
please run the /howtoask command in here 
ivory juniper
/howtoask ffmpg not found how to fix?
vale canyon
nono just run the command, dont type anything else lol
vale canyon
oh haha wait how
verbal patrol
Where can i generate images
rancid gull
My gpu = rtx 5060
Operating system = win 11
Cant hear anything frm the voicechanger
Tut used = https://www.youtube.com/watch?v=SxdnGxicJOg&t=356s
Would appreciate if anyone could help me out
viral mason
My gpu = rtx 5060
Operating system = win 11
Cant hear anything frm the voicechan...
if u used a tutorial u have an outdated voice changerrr
since ur gpu is peak u should get vonovox
zinc steppe
Is it normal for the voice changer to glitch sometimes or is it usually a software or GPU thing
viral mason
Is it normal for the voice changer to glitch sometimes or is it usually a softwa...
depends on what ur doing, gpu, and what voice changer u have
zinc steppe
Oh so it could be a voice character bug?
rancid gull
if u used a tutorial u have an outdated voice changerrr
Alr ty ill try this out
viral mason
Oh so it could be a voice character bug?
Have you tried switching to different voices? Btw if you got the voice changer off YouTube it's outdated
Annnnd, what gpu do u have?
zinc steppe
Have you tried switching to different voices? Btw if you got the voice changer o...
I have the most recent one from the website, I haven't tried multiple voices yet tho
Im not home rn, but i think i have AMD Ryzen 7000
viral mason
I have the most recent one from the website, I haven't tried multiple voices yet...
what?
white shadow
Voice changer isn't outputting my audio. At first when I open the voice changer it works with my headphones right. But when i try using line or cablle to connect it to my game it doesn't work. I have an amd and am using the deltris w-okada changer btw
zinc steppe
what?
Ah sorry, I just mean it wasnt from a youthbe video, I just got everything from the website using local and deltris okada
foggy wagon
@tame oracle can you please help me with the voice modulation software? It distorted in my editing sofware, and it wont even launch, it just gives me an error mesage @tame oracle
nocturne mural
<@1416622683163983902> can you please help me with the voice modulation software...
Someone will come and help you when they have time; no one is obligated to monitor the channel 24/7.
thorn flax
since ur gpu is peak u should get vonovox
you mean the worst possible one??
vonovox is awful
anything you try to do sounds like its in 10 bits and it runs 100% GPU usage constantly
white shadow
anything you try to do sounds like its in 10 bits and it runs 100% GPU usage con...
Its good if you use nvidia
white shadow
welp, i guess thats unlucky
white shadow
i dont have an nvidia so i wouldnt either way
thorn flax
...so why are you saying its good then if you have no idea?
deiteris is the definitive real time voicechanger
white shadow
everyone i know whos used vovonox has said its really good
its why I said unlucky because it seems like a problem only youre dealing with
thorn flax
theyre gaslighting you then
white shadow
ive heard theirs, it sounds good
thorn flax
youll see how horrible it sounds when you use vonovox if you get an nvidia
then theyre not using it lol
white shadow
whatever you say man
thorn flax
plus on top of sounding terrible it runs even the best nvidia gpu to 100% usage conatntly
thorn flax
case in point, vono is so unoptimized its not even funny
viral mason
you mean the worst possible one??
u must not know how to use it correctky
knotty moth
My gpu = rtx 5060
Operating system = win 11
Cant hear anything frm the voicechan...
the tutorial is half year old but it still uses the old version of original wokada 18a that uses older Pytorch version before release date of RTX 50-series gpus, and they don't support it
knotty moth
anything you try to do sounds like its in 10 bits and it runs 100% GPU usage con...
your RTX 2060 is enough for the voice changer alone but don't expect decent gaming performance
otherwise lower the game graphics settings and cap fps to 30 or 60, also please check the voice changer settings as well
btw you can also try tg-develop fork https://docs.aihub.gg/realtime-voice-changer/local/tg-develops-w-okada-fork/
thorn flax
btw you can also try tg-develop fork https://docs.aihub.gg/realtime-voice-change...
That’s what I use cause unlike vonomid it’s actually optimized and I can game and run background processes without any issue
thorn flax
u must not know how to use it correctky
The only way to use it right is to uninstall it
Its honestly unusable
noble walrus
Hello everyone, does anyone know how to render voices using a ready-made voice model in the form of index and pth files? I mean NOT in real time
viral mason
Hello everyone, does anyone know how to render voices using a ready-made voice m...
applio
mellow parrot
does anyone know best setting for deiteris's w-okada fork voice changer that work best in mac mini m2?
i try every setting but the result still not good, is it have anything to do with the model/my microphone?
mortal isle
I tried to use deiteris w-okada rvc and I have it opened but I couldn't connect to it through localhost and instead usded my local IP and now I only see the background of the website. I'm using opensuse tumbleweed. How do I fix it?
warm marlin
No all YouTube tutorials are outdated
and how do it works please 🙂 ?.
mighty quest
does anyone know a good way to integrate an ai player on minecraft java
im using the mineflayer but its just a bot
and i am needing to use an external llm api to control it
but its TOO DUMB
can't even make a simple house of dirt
graceful sun
anyone know why my settings is on gpu 0 but in task manager it says its using a large amount of cpu which is causing the ai to lag
im on a radeon7800xt
which file should i be downloading
<@&1159293204038955078>]
hollow nova
DMR v2 seems to be detected as 40k in applio but it's a 32k pretrain, is there a way to fix this?
low shard
and i am needing to use an external llm api to control it
I haven't really played around with AI bot mods on minecraft java, but what LLM are you currently using?
that seems a funny experiment tho
cold basin
hey i need help
low shard
hey i need help
hello, could you please elaborate?
mighty quest
I haven't really played around with AI bot mods on minecraft java, but what LLM ...
i currently set deepseek-v3.1-terminus
low shard
!howtoask
cold basin
so i set up everything from a youtube video and i cannot hear my voice on anything
i have vega 7 amd ipgu 2gb vram(i know its ass but i wanna test it)
windows11
here is the vid https://www.youtube.com/watch?v=SxdnGxicJOg
rare pecan
on hugging face, where is it to create a voice
cold basin
can someone help me i didnt find anything on the docyment
odd bone
I downloaded the W-Okada voice changer from https://huggingface.co/wok000/vcclient000/tree/main is this a safe site or virus filled shit
odd bone
what gpu do u have
4080
odd bone
it's safe but old
Good to know thanks
viral mason
4080
u should get vonovox since ur gpu is good enough to handle it
form vac lite, run setupx64
and vonovox run setup, then after that is done run start
odd bone
form vac lite, run setupx64
and vonovox run setup, then after that is done run s...
To completely uninstall the old one do I just delete all the files and I'm good?
viral mason
To completely uninstall the old one do I just delete all the files and I'm good?
yea, keep the voice models tho if they're good, just anything in that wokada folder delete with trashcan
burnt veldt
What's the easy way to install applio?
meager coyote
burnt veldt
u should use Vonovox since ur gpu is good enough to handle it ^^
https://github....
there is any video that teach how to configure Vonovox?
viral mason
there is any video that teach how to configure Vonovox?
nope, nobody has made any tutorials, the default settings are usually fine
burnt veldt
nope, nobody has made any tutorials, the default settings are usually fine
i'd like to put a higher delay on voice output to improve the quality, is it possible on Vonovox?
viral mason
i'd like to put a higher delay on voice output to improve the quality, is it pos...
it is yes but delay is bad for conversations
you would do that by increasing block size
viral mason
0.50 is how much in seconds?
I wouldn't know
graceful sun
I haven't really played around with AI bot mods on minecraft java, but what LLM ...
why couldnt i get any help?
<@&1159293204038955078>
i have a radeon 7800xt will it work yes or no
im currently trying to use it and its shit
would love sum type of feedback from the helpers
tame oracle
would love sum type of feedback from the helpers
please read the help guidelines before asking
!howtoask
graceful sun
please read the help guidelines before asking
Bro please just answer my question
Does it work with Radeon cards or no
I’m not doing that entire thing
It’s a simple yes or no
tame oracle
does what work with Radeon cards?
graceful sun
does what work with Radeon cards?
The entire software. When I speak it sounds like shit. Not fully pronouncing the words I’m saying and also using 100% of my cpu. I have a ryzen 7 9800x3d and a Radeon 7800xt
tame oracle
The entire software. When I speak it sounds like shit. Not fully pronouncing the...
so, voice changer then, yes it does work with Radeon cards
graceful sun
so, voice changer then, yes it does work with Radeon cards
So then what could be my issue? I watched a video of someone using it on Roblox and the ai is fully catching everything they are saying. I have a blue yeti mic which should have no problem catching my voice. It cuts my sentences short or just doesn’t even pronounce what I’m saying at all. I can say “hi, how are you doing” the ai will try to say the same thing but fails miserably
tame oracle
So then what could be my issue? I watched a video of someone using it on Roblox ...
so you said you watched a video, did you follow a video tutorial as well?
graceful sun
so you said you watched a video, did you follow a video tutorial as well?
Yes
OK well fuck me and my question then
tame oracle
second guide
viral mason
Impatient people are so annoying:p
gleaming wasp
Good Morning, does anyone know the best website or program for making songs with AI voice?
viral mason
I'd say use applio
It's pretty good
You'll have to separate the vocal and instrumental track tho with uvr
gleaming wasp
I see, perhaps you can help me with this problem in that case
tame oracle
Impatient people are so annoying:p
Welcome to my world
gleaming wasp
I see, perhaps you can help me with this problem in that case
"F:\ApplioV3.2.8-bugfix\env\lib\site-packages\torch\cuda_init_.py:209: UserWarning:
NVIDIA GeForce RTX 5050 with CUDA capability sm_120 is not compatible with the current PyTorch installation.
The current PyTorch install supports CUDA capabilities sm_50 sm_60 sm_61 sm_70 sm_75 sm_80 sm_86 sm_90.
If you want to use the NVIDIA GeForce RTX 5050 GPU with PyTorch, please check the instructions at https://pytorch.org/get-started/locally/"
viral mason
"F:\ApplioV3.2.8-bugfix\env\lib\site-packages\torch\cuda\__init__.py:209: UserWa...
I know nothing about errors as I use applio on Kaggle, a browser and not locally
Maybe a mod or helper could help if they know more than I do
gleaming wasp
I know nothing about errors as I use applio on Kaggle, a browser and not locally
I feel like that would be better Could you tell me how?
viral mason
Using it via kaggle?
gleaming wasp
Yap
viral mason
Sure, give me a minute to get on my pc
gleaming wasp
Thank you very much!
viral mason
Np!
viral mason
Thank you very much!
I'm on my pc now, if u would like to move to dms or just talk here either is fine
gleaming wasp
Yeah thanks you
mortal isle
I tried to use deiteris w-okada rvc and I have it opened but I couldn't connect ...
^
|
mellow parrot
What is the best model for real time voice changer? The pinned link in #🔍│find-models is dead
umbral vapor
i have my chunk on 48000 (1 sec) and extra on 131040 (2.73 sec) but it takes like 30 seconds to catch up on discord. what should i do?
tame oracle
What is the best model for real time voice changer? The pinned link in <#1163592...
there isn't just one "best" model for it
tame oracle
i have my chunk on 48000 (1 sec) and extra on 131040 (2.73 sec) but it takes lik...
please read the help guidelines before asking
!howtoask
white shadow
can someone lmk what the best working version of w-okada works for amd
im not too up to date with anything anymore 😭
odd shale
can someone lmk what the best working version of w-okada works for amd
Don't use w okada anymore
Go for vonovox instead
white shadow
does that support amd now?
white shadow
yeah it only works for nvidia
torpid nacelle
hello
torpid nacelle
someone here ? @viral mason
nocturne mural
sorry im new
knotty moth
thank u for teaching how to ask
as it said, you should explain the problem first
hallow thistle
OK well fuck me and my question then
If I could scold you either way, I just won't. AMD Radeon RX 7800 XT will work with Deiteris/Tg Develop W-Okada fork. The thing is the one you got from tutorial video from YouTube is outdated, where its DirectML variant is also bugged. 
hallow thistle
thank u for teaching how to ask
No excuse. If you need help about something, basically explain about your issue, the program you wanna use/you tried, and your PC specs (like GPU). You wouldn't simply say "hello" and expect one to answer or ask you back in a help channel.
stray copper
is refinegan in any applio local build?
idle shell
it gives me "typeerror failed to fetch" whenever i try to use okada whys it doing that
knotty moth
<@681186927151546397>
use reply instead of pinging
knotty moth
doesn’t ngrok suck ass now
I said another tunnel otherwise
it's just that ngrok has limited requests/min to allow but you can still run applio anyway
idle shell
it gives me "typeerror failed to fetch" whenever i try to use okada whys it doing that
paper lotus
I have a ryzen 5 5700 rx 6800 oc 16gb and 32gb ram 1tb ssd and 3tb hdd can i run this on marvel rivals
1440p max settings
cunning gyro
hi
stray copper
I said another tunnel otherwise
horizon also doesn't work with filebrowser
vestal geyser
what is the best ai to use for helping you study for school?
wild oyster
burnt veldt
Why do my exported models in Applio have robotic voices? I configured the training with 1200 epochs and 10 good quality audio tracks. Am I doing something wrong?
I'm using the version 3.6.0 of Applio
unborn briar
Can you please tell me what to do?
the sound is not output for some reason
I even reinstalled the cable.
There is no sound output to the discord, who can help with private messages?
turbid isle
Why do my exported models in Applio have robotic voices? I configured the traini...
When u trained your model did u check the overtraining detector box? I usually do that
My model isn't that good but the result is acceptable
unborn briar
kto russkiy i mojet pomoch?
turbid isle
There is no sound output to the discord, who can help with private messages?
Надо что то поменять
Я не русский, хз язык ваш
burnt veldt
When u trained your model did u check the overtraining detector box? I usually d...
yes, it was disabled, i trying other things to see if improve it
turbid isle
yes, it was disabled, i trying other things to see if improve it
Oh damn, I always enable it
Try to enable it and use a smaller epoch to see how it work like maybe 100-200 Then u can continue to train for higher epoch level
burnt veldt
Ok
burnt veldt
Thx man
i think the other problem that i'm having is this, i'm trying to train Brazillian voices, trying to figure out how to add the Nanashi pretrained model
turbid isle
ahhh idk about those, i use hifi gan in default
burnt veldt
👨🏭
hallow thistle
what is the best realtime voicechanger
W-Okada voice changer. The user you replied to mentions about "Vonovox", a complete alternative to W-Okada that also uses RVC voice models but with different interfaces.
cerulean pond
i wan't very clean interface and ultra realtime voice changer
hallow thistle
Make sure to read help guidelines before start going outside Discord server, looking for an "outdated" W-Okada version from YouTube tutorial videos.
hallow thistle
i wan't very clean interface and ultra realtime voice changer
Want, wantn't or wasn't?
By the way, what are your PC GPU and operation system? Because a voice changer program often performs better with a discrete GPU.
hallow thistle
what is the best ai to use for helping you study for school?
The only best AI to help study in school is you. There are also some services that offer similar usages, but they won't be the same level as you. 
azure charm
does anyone know why the delay with the app is so long for me? the audios really choppy and takes ages to go through the delay literally says 200 seconds, my specs aren't that bad i have a mid range pc with a ryzen 5 5500x and a intel arc a380
steady hinge
Could anyone tell me why when I use the Realtime Voice Changer Client it doesn't change my voice at all?
<@&1159293204038955078>
Like even if I raise the pitch it does nothing..
hallow thistle
does anyone know why the delay with the app is so long for me? the audios really...
hallow thistle
Could anyone tell me why when I use the Realtime Voice Changer Client it doesn't...
Make sure to read help guidelines before start asking anything. What is your PC GPU? And did you follow any tutorial or guide before this?
steady hinge
Make sure to read help guidelines before start asking anything. What is your PC ...
I have a 5090 and Ryzen 7 9800x3d and yeah I followed like 3 tutorials online and downloaded multiple versions but every time my voice just stays the same.
hallow thistle
I have a 5090 and Ryzen 7 9800x3d and yeah I followed like 3 tutorials online an...
Try Vonovox. https://docs.aihub.gg/realtime-voice-changer/local/vonovox/ https://github.com/dr87/Vonovox/releases/tag/v1.6.9
hallow thistle
Vonovox is a complete alternative to all other W-Okada versions. Both use RVC voice models, but Vonovox has a whole different UI.
hallow thistle
You're welcome. 
steady hinge
Why wasn't w-okada working for me specifically though?
hallow thistle
The W-Okada version you used might be an original and outdated one. There's Tg Develop W-Okada fork (b2397), this one is better than the original W-Okada, although Vonovox is often preferred for audio quality.
steady hinge
oh ok
hallow thistle
The original W-Okada didn't make one to work properly with NVIDIA GeForce RTX 50 GPU, so it kind of happens.
steady hinge
ohh that makes sense now
becuase it did come out before 50 series came out.
Now when I talk it stutters really bad 😭
wait
turbid isle
Sooo how to make your voice model be able to whisper
Cuz whenever I try to whisper its just meh
But I can do high deep evil voice
mystic sonnet
all of my voices sound the same
candid badge
hi guys is there a free alternative for virtual audio cable ? it keeps saying trial again and again
mystic sonnet
my audio is very grainy
burnt veldt
Does anyone know how do i increase delay time on Vonovox? To improve voice quality
mystic sonnet
how do i fix my shi having a lisp
fossil sage
What's the best audio upscaling
hollow mantle
Hello, I wanna ask which realtime AI voice conversation I should get
Specs:
16 GB of RAM
CPU: AMD Ryzen 5 8645HS
GPU: NVIDIA GeForce RTX 3050 - 6 GB
Can anybody link me to a version that's smooth and not choppy? I heard RVC is good
stray remnant
What voicechanger is the best and free?
viral mason
What voicechanger is the best and free?
Depends on ur gpu
viral mason
Hello, I wanna ask which realtime AI voice conversation I should get
Specs:
16 G...
You should try out Vonovox, it's the current best and works only with Nvidia
hollow mantle
You should try out Vonovox, it's the current best and works only with Nvidia
Isn't Vonovox for high end rtxs
I heard 3050 is basically "the bare minimum"
viral mason
I mean it could work, but if u don't want the risk of it not working too well u could also try wokada tg fork
It's what I currently use while Vonovox's current beta gets an update to a stable version
stray remnant
Depends on ur gpu
I have a 1660 super with ryzen 7 5600x 8 cores with 32gb ddr4 ram
viral mason
I have a 1660 super with ryzen 7 5600x 8 cores with 32gb ddr4 ram
At best you'd be able to use w-Okada deiteris unless you use your AMD card which u could use wokada tg fork
-rt
stray remnant
Where can I download that?
viral mason
Wokada tg fork is the best of the two I mentioned, you can download them from the guide that is associated with each one
dusky lark
whenever i say something 1 second later it repeats itself how do i fix
hollow mantle
I mean it could work, but if u don't want the risk of it not working too well u ...
Mmmmm I need a 100% confirmation that 3050 is good, you said Vonovox would work but it works only with NVIDIA which is what I have
Can you give me a link to both? And if anyone here has a clear answer about 3050 that would be awesome, I haven't actually bought the laptop with these specs yet but I am going to soon, that's why I am mostly just asking
dusky lark
@hollow mantle
do you know how i can fix this
whenever i say something 1 second later it repeats itself
hollow mantle
It's supposed to do that
viral mason
Mmmmm I need a 100% confirmation that 3050 is good, you said Vonovox would work ...
Sure! Just one sec
hollow mantle
You press stop
dusky lark
ohh i get it
hollow mantle
Sure! Just one sec
Do they at least have spec requirements I can read? Maybe 3050 is good enough
viral mason
Mmmmm I need a 100% confirmation that 3050 is good, you said Vonovox would work ...
Vonovox: https://github.com/dr87/Vonovox/releases/tag/v1.6.9
https://docs.aihub.gg/realtime-voice-changer/local/vonovox/
Wokada tg fork:
https://github.com/tg-develop/voice-changer/releases/download/b2397/voice-changer-windows-amd64-cuda.zip.001
https://docs.aihub.gg/realtime-voice-changer/local/deiteris-w-okada-fork/
I put the guides with the downloads
hollow mantle
Alrighty ty, any voice model from #1175430844685484042 would work?
viral mason
Any of them work yea, only ones that won't work are the refinegan models I made with spinv2 recently
Homer, glados, and bugs bunny
Those only work in the most recent beta of Vonovox
hollow mantle
Gotcha, and Vonovox does work, the minimum is far lower than I expected
Oh this looks kinda hard, is there a YouTube tutorial on how to set everything up?
viral mason
Gotcha, and Vonovox does work, the minimum is far lower than I expected
What was the minimum?
viral mason
Oh this looks kinda hard, is there a YouTube tutorial on how to set everything u...
All u do is run setup, and then run start
For Vonovox at least
The settings that are best are default actually, but if u want less delay decrease block size
Btw do u have vac lite or vb cable
U need a virtual audio cable to use this in games
hot lagoon
i dont know how to find the anvuew v2 for uvr
like what yaml and cpk I need to used
idle shell
it gives me "typeerror failed to fetch" whenever i try to use okada whys it doing that
past thicket
ive done all the settings right but it just isnt working
tame oracle
it gives me "typeerror failed to fetch" whenever i try to use okada whys it doin...
please read the help guidelines before asking
tame oracle
ive done all the settings right but it just isnt working
please read the help guidelines before asking
!howtoask
knotty moth
Mmmmm I need a 100% confirmation that 3050 is good, you said Vonovox would work ...
even though 3050 laptop 4 GB can still run the voice changer alone, I'd rather suggest the one with 8 GB vram or better 12 GB to be able to run some modern games
teal geyser
im getting an error "Voice Changer is not selected." while using MMVC okada web client
misty glacier
how do i use the voice changer
long oyster
I don't know what's wrong with my w odaka app ( RuntimeError: CUDA error: no kernel image is available for execution on the device
CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.
For debugging consider passing CUDA_LAUNCH_BLOCKING=1
Compile with TORCH_USE_CUDA_DSA to enable device-side assertions.)
wait i need to download gtx 5000? right guys?
knotty moth
wait i need to download gtx 5000? right guys?
either downgrade to an RTX 40-series gpu or get vonovox/tg-develop fork as listed below
-rt
hollow mantle
even though 3050 laptop 4 GB can still run the voice changer alone, I'd rather s...
Those are the specs I can currently afford. You said 3050 4gb can still run, it's 6GB.
Are you saying Vonoxo can run with these specs just fine with at least discord involved? I am not planning on playing huge modern games atm
knotty moth
Those are the specs I can currently afford. You said 3050 4gb can still run, it'...
isn't the 6 GB variant desktop one?
knotty moth
Those are the specs I can currently afford. You said 3050 4gb can still run, it'...
discord shouldn't take noticable GPU load unlike games, and could say so for vtube studio
hollow mantle
discord shouldn't take noticable GPU load unlike games, and could say so for vtu...
Not sure if it's a variant one, I just wanna ensure I don't sound like this lmao https://m.youtube.com/watch?v=TR8Tdfj6_-c
Something smooth and actually sounds normal
The model is a variant with better components for mid-tier gaming
knotty moth
Not sure if it's a variant one, I just wanna ensure I don't sound like this lmao...
it's more likely the model quality
hollow mantle
Not sure if it's a variant one, I just wanna ensure I don't sound like this lmao...
This voice is from RVC Okada, the person in this video has a rtx 4060 ti16 GB and a i10100f
hollow mantle
it's more likely the model quality
It's probably an ASMR model and that's what made it not as good
hallow thistle
If you have NVIDIA GeForce RTX GPU, there's Vonovox. 
hollow mantle
I know, I am planning on using it and thankfully the requirements are below with what I am going for based on the doc https://docs.aihub.gg/realtime-voice-changer/local/vonovox/
hallow thistle
wait i need to download gtx 5000? right guys?
There's no NVIDIA GeForce "GTX" 50; there's RTX 50. For GeForce RTX 50 GPU, there are Tg Develop's W-Okada fork and Vonovox.
low aspen
Hello, I have a question for you. Are you able to create your own game scripts? I need a script for the game Case Paradise. I'll send you a link to the game, but I can't find the script anywhere. It's not a very popular game. I'm interested in either item spawning or duplication. Thanks in advance.
knotty moth
It's probably an ASMR model and that's what made it not as good
there could be some good enough ones, it's more of the dataset source quality and how properly the model maker cleaned & prepared it
hollow mantle
Do you have a recommendation on one that's good?
hallow thistle
hollow mantle
hallow thistle
If you could find one, there are like over thousands of voice models.
hallow thistle
how do i use the voice changer
The Tg Develop's W-Okada fork, double click MMVCServerSIO.exe. Asides that, make sure to read help guidelines before start asking anything than a simple query.
knotty moth
There are a lot
try at least listen to the samples, otherwise gotta have to try the model
hallow thistle
it gives me "typeerror failed to fetch" whenever i try to use okada whys it doin...
What is your PC GPU? Did you follow any tutorial or guide before this? Because the original and outdated W-Okada version often found in tutorial videos from YouTube.
hollow mantle
warm yoke
Heyy y’all, my friend want to do music with ai voice model, and he’s asking for a pc setup to do that can someone help me please ?🙏🏽
hallow thistle
Heyy y’all, my friend want to do music with ai voice model, and he’s asking for ...
!howtoask
warm yoke
Oops sorry
hallow thistle
To check his PC's specs, tell him to open Task Manager, go to Performance tab, and spot GPU 0 or GPU 1. If you mean what's the best specs for a PC to run "Applio RVC fork" locally, you simply explain about your query.
warm yoke
He have no computer yet 😅
idle shell
What is your PC GPU? Did you follow any tutorial or guide before this? Because t...
i have a 4070ti and i followed a tutorialyeah
hallow thistle
i have a 4070ti and i followed a tutorialyeah
Try Vonovox. https://github.com/dr87/Vonovox/releases/tag/v1.6.9
hallow thistle
He have no computer yet 😅
So is it simply that? You can ask something like "what's the best PCs specs (like GPU) to run Applio RVC locally?". For a decent PC that can also be used to play some games and run AI tasks, tenth generation Intel Core and NVIDIA GeForce RTX 3060 16GB can be enough, although there are newer Intel Core / Core Ultra + RTX 40 or RTX 50 if your friend wants more performance and budget is possible. 
As much as I hinted about potential PC specs, you can also do some research about best working PC for your needs, that doesn't have always be exact as my statement.
warm yoke
Thanks Lucy, I appreciate
hallow thistle
You're welcome.
mystic tree
Hello, I have a question.
I've read the guide and my wokada works perfectly without any issues. Is it possible to connect wokada from my PC to my smartphone because I want to play games on it and use the voice changer.
Sorry if my English isn't very good because I'm using a translator.
Thank you
paper wadi
whats a good epoch amount for training a voice model?
lunar belfry
-rvc
lunar belfry
-colab
safe glacier
i need some help with a error with applio it says: RuntimeError: Failed to import transformers.models.hubert.modeling_hubert because of the following error
operator torchvision: :nms does not exist
i have rtx 3050 laptop gpu
safe glacier
i need some help with a error with applio it says: RuntimeError: Failed to impo...
i downloaded the precompiled 3.6.0 applio
latent knoll
i got a .index file, a .pth file, and a .npy file, what do i do with them?
simple ore
i need some help with a error with applio it says: RuntimeError: Failed to impo...
it seems like a mismatch of some libraries
simple ore
i got a .index file, a .pth file, and a .npy file, what do i do with them?
mode.pth and added index are used, .npy is a temp file
latent knoll
mode.pth and added index are used, .npy is a temp file
how do i use the files tho where do i put them
simple ore
how do i use the files tho where do i put them
depends on the application you're trying to use
latent knoll
depends on the application you're trying to use
idk man i wanna put drake vocals on a lil uzi vert song
valid scarab
I just downloaded the latest cuda version of w-okada. I have a 5060 ti, and it says it's sm_130. But it says "The current PyTorch install supports CUDA capabilities sm_37 sm_50 sm_60 sm_61 sm_70 sm_75 sm_80 sm_86 sm_90 compute_37." I don't know what PsTorch is, how it relates to w-okada, or how to fix this, any help?
simple ore
I just downloaded the latest cuda version of w-okada. I have a 5060 ti, and it s...
-rt
simple ore
1st two links
viral mason
if u got ur voice changer from a youtube tut or if the links say demo client they're super old
valid scarab
yeah i got it from w-okadas github hugging face
simple ore
i downloaded the precompiled 3.6.0 applio
possibly you did not unzip it properly, i'll give it a test in a sec
viral mason
I'd say choose between vonovox or wokada tg fork
simple ore
i downloaded the precompiled 3.6.0 applio
just gave it a quick test with inference, looks fine to me
valid scarab
if u got ur voice changer from a youtube tut or if the links say demo client the...
do you know any programs that are more mainstream?
i cant find anything online about either of those 2
safe glacier
possibly you did not unzip it properly, i'll give it a test in a sec
i used winrar. maybe if i try with 7-Zip?
viral mason
do you know any programs that are more mainstream?
The two I recommended are the only two that are good, just read the guide provided for the one you want to try.
Anything online will be either really outdated and old or pay to use
For Vonovox after you extract the file run setup, then run start
For wokada tg fork after extraction just run mmvcserversio
simple ore
i used winrar. maybe if i try with 7-Zip?
maybe, make sure you have enough free space
safe glacier
maybe, make sure you have enough free space
same error
i have enough free space
maybe its because a wrong cuda version of torch?
simple ore
maybe its because a wrong cuda version of torch?
the compiled version comes with all required libraries and python
safe glacier
the compiled version comes with all required libraries and python
im in windows 10 21H2
safe glacier
do you have a standalone cuda toolkit installed?
yes
simple ore
it usually does not cause issues, but who knows...
digital bluff
y'all is there a site to generate ai voices for free im not paying weights for this it used to be free
or train them whatever
safe glacier
it usually does not cause issues, but who knows...
so i need to uninstall?
digital bluff
soo do u know anyy
safe glacier
do you have a standalone cuda toolkit installed?
i checked my storage and i only have the drivers
obtuse igloo
so im trying vonovox now since the other one is outdated but it says model not properly initialized and its just warming up the voice conversion
simple ore
so i need to uninstall?
you can try
obtuse igloo
what do i do guys?
viral mason
so im trying vonovox now since the other one is outdated but it says model not p...
What gpu do u have?
It only works with Nvidia:3
viral mason
Ok 👍
obtuse igloo
so what should i do?
viral mason
Try closing Vonovox and reopening it to see if it happens again
obtuse igloo
i did that already
heavy cypress
is there like a tut for ai vc on mac
safe glacier
you can try
using Miniconda Prompt to activate the conda env of the precompiled applio and using this command "pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118"
fixed the problem, maybe you want to add this command to the guide if anyone have this problem too
simple ore
using Miniconda Prompt to activate the conda env of the precompiled applio and u...
i dunno why would you go with cu118, but okay
also you did not need to activate, you could just run env\python -m pip uninstall ... / env\python -m pip install
safe glacier
also you did not need to activate, you could just run env\python -m pip uninstal...
oh okay thank you
wheat socket
I got the site thing to work but it wont pick up the mic to discord
solar frost
hello
rigid yoke
I need help to find TTS or model for Wukong and Pigsy which use for "what kind of (X) is this/what kind of awesome is this" i found some website but it was paid https://www.reddi[.]com/r/HelpMeFind/comments/1lyv7ff/help_me_find_the_tts_for_the_what_kind_of_x_is/
knotty moth
using Miniconda Prompt to activate the conda env of the precompiled applio and u...
that cuda version is old, only okay for GTX 10-series
safe glacier
that cuda version is old, only okay for GTX 10-series
then why these error happens with the cuda version of the precompiled zip?
i typed the command to see the cuda version and it shows cuda 13.0
knotty moth
then why these error happens with the cuda version of the precompiled zip?
i havent seen your gpu but I suppose GTX 10-series?
the latest cuda 13 needs 20-series/newer
safe glacier
i havent seen your gpu but I suppose GTX 10-series?
rtx 3050 laptop gpu
zenith scaffold
can someone help me to make a chat bot?
that i can host for free :D
i have 6gb vram tho and 2060 gpu
:/
jade moth
okada
tame oracle
saying "okada" doesn't mean anything or prompt us to do something, if you need help finding a voice changer for you id suggest you read the help guidelines and ask
cyan mantle
Hi ! Hope ur all going great ! 😃
I wonder how to make a voice model, I have some minutes of a voice I like (worms voice lmao) and I would like to speech some text using this. How can I create a model please ?
viral mason
Me? 
cyan mantle
Me? <:ruru_shocked:1420423263414386688>
Hahaha 😂
simple ore
Hi ! Hope ur all going great ! 😃
I wonder how to make a voice model, I have so...
you dont need a voice model for tts, you can just use a zero-shot tts that uses 10 seconds of audio
cyan mantle
you dont need a voice model for tts, you can just use a zero-shot tts that uses ...
I'm still going to create a model I think. Because I already have excerpts from the voice of the Worms, and a good bunch of lines that I would like to create with the voice
pulsar osprey
how to use voice-models?
turbid isle
wait how can i train a model in russian?
kindred ledge
i cant use my gpu, when i set to gpu, RVC error
when i choose gpu
tired minnow
oh right
hollow mantle
How big is Microsoft Visual C++ Redistributable Package
hallow thistle
How big is Microsoft Visual C++ Redistributable Package
Megabytes.
hallow thistle
i cant use my gpu, when i set to gpu, RVC error
Try Vonovox instead of Deiteris W-Okada (b2332). That b2332 one didn't make to run with GeForce RTX 50 series GPU. https://github.com/dr87/Vonovox/releases/tag/v1.6.9
If you'd like to stay with W-Okada for easier interface, there's 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
kindred ledge
oh okay, then i download this 3 files?
hallow thistle
oh okay, then i download this 3 files?
voice-changer-windows-amd64-cuda.zip.001 and voice-changer-windows-amd64-cuda.zip.002. Not voice-changer-windows-amd64-cpu.zip.
alpine drift
Hi is there any video for creat a rvc voice model?
hallow thistle
Hi is there any video for creat a rvc voice model?
If you're looking for a tutorial video from YouTube that sometimes tells outdated information, I'd rather not recommended. Instead, there's a guide for "Applio RVC fork".
-rvc
alpine drift
Ty
kindred ledge
okay thankss
empty flint
I'm using Applio and I am getting a weird robotic background noise on a model that used to work perfectly.
- Same model, same settings as before.
I even tried to redo an old recoding I had with the same setting and got a very different and much worse result.
all I did was sleep? and I woke up to my Applio just acting up
hallow thistle
I'm using Applio and I am getting a weird robotic background noise on a model th...
When you say you "were sleep" and Applio RVC acting up, it sounds like some creepypasta storyline where the program changes its settings while you were unaware of it. 
empty flint
When you say you "were sleep" and Applio RVC acting up, it sounds like some cree...
I wish it was that but no, SAME settings SAME everything and all of a sudden the output is trash
hallow thistle
Make sure to use latest version of Applio RVC, review and redo your exact steps, and use your headphones to hear them closely again.
empty flint
Using 3.6.0
Headphones are full blast in my skull
I even wore the same clothes I was wearing yesterday and still it sounds so weird
hallow thistle
If that issue has a possible explanation, it could be the voice model itself being in low quality, the audio has background noise, or you had a phenomenon called "Mandela effect", where you think you remember the initial generated audio once sounded good but it sounded terrible in reality as more you listened to it.
If you have these audio files (both first and last one) saved to your drive, you can try send them to here to make sure the issue actually persists.
hallow thistle
rvc webgui works fine on amd 9070 xt card?
Applio RVC fork is yes, more preferred. But for the mainline RVC GUI (like RVC1006AMD_Intel.7z), this one is outdated.
Check out this one. https://docs.aihub.gg/rvc/local/applio/#amd-on-windows-precompiled-fix
lunar belfry
If ur gonna train a model use Kaggle applio or use local applio
i know this was tomorrow but i used the first one (applio) to train an alphabet lore model
viral mason
Wdym it was tomorrow
lunar belfry
beacuse it shows right closer on your username
it says "Ontem às 14:20" sorry if its portuguese
viral mason
Idk what 14 o clock is as I use the 12 hour clock in America
But I think as long as you used applio for training a model it's probably fine
simple schooner
my w-okada voice changer is bad with my amd rx 7800xt 16gb vram
viral mason
my w-okada voice changer is bad with my amd rx 7800xt 16gb vram
Did you get it off a YouTube tutorial
viral mason
Idk what direct ml is so I'm assuming that's super old and outdated
viral mason
Yep, follow the guide and download the correct version for amd, I wouldn't know what the AMD version is but someone like Namari could
simple schooner
oke
viral mason
Since I use Nvidia I only know the correct files to download forNvidia
If u have questions on how to use it tho I know what to do there
green sparrow
hey yall, i have a question, i'm very new to the ai-voice changer scene, and i'm interested in potentially using w okada's voice changer to start streaming using my OC robot's voice, since she uses an old text-to-speech engine as her voice. What would be the best way to learn how to train my own voice model on clips I can generate using the old TTS engine?
I've no idea at all where to even get started with learning how to train my own model so any help would be very much appreciated! thank you :)
lunar belfry
"Not enough data present in the training set. Perhaps you forgot to slice the audio files in preprocess?"
BRO
WHAT IS THIS ERROR
MY FILES ARE ALREADY SMALL AND ITS STILL SAYING THIS
STUPID COLAB
AND IT STOPS AT STOP TRAINING
I CANT TRAIN IT AGAIN
AND IM USING APPLIO TO MAKE IT AND ITS ALREADY SLICED
BUT ITS STILL MAKING ERROR
im still getting the SAME ERROR GRAAAAAAAAAH
viral mason
hey yall, i have a question, i'm very new to the ai-voice changer scene, and i'm...
-rt
viral mason
it's the second guide
I don't have amd so I wouldn't know exactly what files to download but they are probably in the guide
green sparrow
and will this help me to train my own model to use in that wokada fork you suggested i use?
viral mason
and will this help me to train my own model to use in that wokada fork you sugge...
there's actually a seperate thing for that
wokada tg fork is only the voice changer
-rvc
green sparrow
ah okay, that makes sense
viral mason
applio is the current most used voice cloning software
green sparrow
and the applio docs have info about training an rvc model for use within wokada?
lunar belfry
And it ignored me
green sparrow
sorry i'm asking so many questions i'm jumping in at the deep end lmao
safe glacier
that cuda version is old, only okay for GTX 10-series
okay, i changed the "118" to "121" and now works fine
viral mason
and the applio docs have info about training an rvc model for use within wokada?
yes
the model u make with the audio can be used in both wokada and other places that use rvc
like for covers on weights or in applio itself
green sparrow
that's perfect thank you so much 
viral mason
you're welcome ^^
btw if u have questions u can ask me, if I can't answer them maybe a mod or helper can
viral mason
WHAT IS THIS ERROR
I have gotten this before too, idk how to fix it or what causes it
woeful cobalt
Could I self-host a small LLM? I have an AMD Ryzen 7 5700G, a Radeon RX 6600 XT with 8 GB of VRAM, and 16 gigabytes of DDR4
simple schooner
@viral mason i just installed tg develops what settings do i do?
or is there a tut?
viral mason
<@1023278814752677918> i just installed tg develops what settings do i do?
did u run mmvcsersio?
viral mason
do u have vac lite?
viral mason
it's a virtual audio cable that would connect the voice changer with other software like discord or games in general
simple schooner
it's a virtual audio cable that would connect the voice changer with other softw...
not yet but i will be downloading
radiant cypress
Is there any reverse search tool
I got some girl dming me stuff on insta
Cant tell if she is using a fake id or not
Tried google lens but ntg came up
tame oracle
Is there any reverse search tool
TinEye is a reverse image search tool
radiant cypress
TinEye is a reverse image search tool
Tried but it didnt work
tame oracle
Tried but it didnt work
i really dont know anymore then
radiant cypress
Welp nvm it ig
safe glacier
Could I self-host a small LLM? I have an AMD Ryzen 7 5700G, a Radeon RX 6600 XT ...
just use kobold
is the easiest way
knotty moth
Could I self-host a small LLM? I have an AMD Ryzen 7 5700G, a Radeon RX 6600 XT ...
LM studio
but given such limited vram, it'd be able to run < 10b llms
teal geyser
im using w-okada fork and for some reason, when I have my input/output set correctly, it occasionally plays on my browser tab and not through my vb cable
ok i fixed by selecting server and then client
simple schooner
how do i make voice changer work in dc?
teal geyser
anytime i select a different voice model, i get "An error occured during voice conversion..." but it still works, is there a reason or anything I should do to fix this?cmd 2025-12-09 19:03:44,930 INFO [RVCr2] Initialized. 2025-12-09 19:03:44,951 ERROR [VoiceChangerManager] The expanded size of the tensor (9600) must match the existing size (7431) at non-singleton dimension 0. Target sizes: [9600]. Tensor sizes: [7431] Traceback (most recent call last): File "voice_changer\VoiceChangerManager.py", line 212, in change_voice audio, vol, perf = self.vc.on_request(receivedData) File "torch\utils\_contextlib.py", line 116, in decorate_context return func(*args, **kwargs) File "voice_changer\VoiceChangerV2.py", line 162, in on_request result, vol = self.process_audio(audio_in) File "voice_changer\VoiceChangerV2.py", line 152, in process_audio self.sola_buffer[:] = audio[block_size : block_size + self.crossfade_frame] RuntimeError: The expanded size of the tensor (9600) must match the existing size (7431) at non-singleton dimension 0. Target sizes: [9600]. Tensor sizes: [7431]
dark frost
hi guys, i installed vb audio virtual cable for my voice changer bc i saw that in a tutorial but now i cant hear any sound from my computer
opal depot
Hello. Im using kaggle cloud RVC, and when I start server, this is the error I get.
2025-12-10 01:27:41,266 ERROR [VoiceChangerManager] Input type (torch.cuda.FloatTensor) and weight type (torch.FloatTensor) should be the same
untold cradle
hi guys, i installed vb audio virtual cable for my voice changer bc i saw that i...
what audio cable are you using? like vb
dark frost
i was able to fix it
forgot to say
but
my thing is hella laggy
like
the voice changer has a pretty high delay and its always cutting like the sound
im using the best settings for it
untold cradle
i was able to fix it
dm me
dark frost
can smb dm me to help me figure out why my voice changer has such a delay
viral mason
quasi condor
YOOO IM INTERESTED SIGN ME UP MY GUY
opal depot
Hello. Im using kaggle cloud RVC, and when I start server, this is the error I g...
lightning.ai also seems bugged idk what's wrong please help
opal depot
it works fine
I cant post images here to show
but on lightning.ai it says
[SIO] rconnection failed Error: xhr post error
over and over again
on kaggle it says
2025-12-10 01:27:41,266 ERROR [VoiceChangerManager] Input type (torch.cuda.FloatTensor) and weight type (torch.FloatTensor) should be the same
i changed nothing in the code
quasi condor
I cant post images here to show
are you trying to open applio?
quasi condor
what is applio
or wokada
knotty moth
what is applio
it is RVC inference application, but the latest version (3.6.0) has realtime feature. sorry that it and the cloud wokada are outside my knowledge tho.
opal depot
https://cdn.discordapp.com/attachments/427903255096459294/1439592941931724810/im...
Rice Cooker....
wicked sorrel
my mic isn't being picked up at all, i have it set to my regular mic and the output to my virtual cable
tame oracle
!howtoask
hallow thistle
my mic isn't being picked up at all, i have it set to my regular mic and the out...
Make sure to read help guidelines before start asking about W-Okada realtime voice changer. Check your microphone. What is your PC GPU? And did you follow any tutorial or guide before your query?
wicked sorrel
Sorry I fixed it thanks
Just using old models
Switched up to other ones and they worked fine
hallow thistle
That doesn't answer anything.
wicked sorrel
Fine
I checked my microphone, was being picked up. Have rtx 3050ti laptop
And I followed a guide yes but it's fine now
hallow thistle
I'm more of going into much deeper issue, not a short-term solution as what you stated.
hallow thistle
If you have NVIDIA GeForce RTX GPU, 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 If you have followed any tutorial from YouTube, that one likely telling you to use an outdated version (v.1.5.3.18a).
wicked sorrel
I'll reinstall tomorrow and see if the older models work then thanks
I think the guy said that if you have amd to download a different github version (I did) but now I'm forgetting if he meant cpu or gpu
I have ryzen 5
hallow thistle
AMD Ryzen is CPU, while AMD Radeon RX is a dedicated GPU. Your laptop has RTX 3050 Ti, so better use a version that compiled to run with NVIDIA GPU. The "DirectML" variant of W-Okada can also work with NVIDIA GPU, not just AMD/Intel, but not always recommended.
wicked sorrel
Gotcha
I hear there's another version of the software (or different software?)
What's the big difference
hallow thistle
There's Vonovox, a complete alternative to other W-Okada versions. There's big difference about UIs, settings, though they both use RVC voice models and all work similar.
hallow thistle
what is applio
Applio RVC (Retrieval-based Voice Conversion) fork. The program is not realtime voice changer the way W-Okada voice changer does, Applio RVC is used to do AI cover and even train an RVC voice model.
Either way, RVC doesn't always mean realtime voice changer, that said.
hallow thistle
<@&1159293204038955078> get in here im not a helper
hallow thistle
Hello. Im using kaggle cloud RVC, and when I start server, this is the error I g...
Now, delete your current MMVCServerSIO (voice changer) Kaggle notebook and try follow this guide. https://docs.aihub.gg/realtime-voice-changer/cloud/tg-develops-w-okada-fork-kaggle/
hallow thistle
anytime i select a different voice model, i get "An error occured during voice c...
Kaggle, Google Colab or you run locally?
hallow thistle
Could I self-host a small LLM? I have an AMD Ryzen 7 5700G, a Radeon RX 6600 XT ...
8 GB VRAM is a bit small by today's standard, although running smaller LLM model is possible if the model is optimized to use under 7.9 GB of GPU VRAM and work with Radeon RX GPU especially. For more VRAM, there's NVIDIA GeForce RTX 3060 12GB as a starting point.
hallow thistle
And it ignored me
If it possible, you can use @ helper role ping to call some other helpers to investigate your issue. When you rant a small rant in caps lock or accuse other people for ignoring you, anyone can either wait you to calm down or won't talk to you after all.
woeful cobalt
8 GB VRAM is a bit small by today's standard, although running smaller LLM model...
I'd prefer to stay away from Nvidia where possible largely because I don't like the company and because I use Linux
Nvidia, famously and historically, provided rather poor driver support from Linux
hallow thistle
I'd prefer to stay away from Nvidia where possible largely because I don't like ...
Your opinion, your choice.
While I use Linux distro for some parts, like trying to compile FFmpeg from start, I still struggle to get compiler going without having to do sudo apt-get or install some packages needed/required (like libfdk-aac and libx265). I'm more interested in NVIDIA GPUs, but if you could find AMD Radeon RX that's comparable to a specific high-performance NVIDIA GPU you could've done that, same goes to Intel too.
woeful cobalt
I see
woeful cobalt
While I use Linux distro for some parts, like trying to compile FFmpeg from star...
Debian, I take it?
hallow thistle
Debian, I take it?
I once tried Debian on Oracle Virtualbox, but I forgot to check GUI install and connect the internet at installation, and the result was that it shown as text mode-only terminal. For more consumer-friendly uses, there's Ubuntu, while Debian leans a bit at professional. There are some Linux distros available, but Debian and Ubuntu are what I know.
woeful cobalt
I once tried Debian on Oracle Virtualbox, but I forgot to check GUI install and ...
I feel like those mistakes regarding Debian are on you
hallow thistle
I feel like those mistakes regarding Debian are on you
Yes, these were my mistakes. 
woeful cobalt
Yes, these were my mistakes. <:Marisa_DekaThink:870814375232045066>
Yeah
I can definitely still recommend Debian though
knotty moth
Your opinion, your choice. <:TetoShrug:1243301470430826567>
nah it's true that the Nvidia driver support is worse than AMD one in linux
knotty moth
I can definitely still recommend Debian though
ubuntu and arch/derivatives aren't bad either
woeful cobalt
ubuntu and arch/derivatives aren't bad either
Arch itself isn't bad either, if a little complicated
I use Arch myself
azure patio
im lost. where to train a model, applio kaggle gave a model format (512, 256, 1) when it shouldve been (256,256) or smth
knotty moth
im lost. where to train a model, applio kaggle gave a model format (512, 256, 1)...
it seems more likely mismatched sample rate of the preprocessed model with the pretrain and/or the vocoder (hifigan or refinegan).
to be more clear, show screenshot of the configurations (ask me for image perms first).
azure patio
it seems more likely mismatched sample rate of the preprocessed model with the p...
i used 40k sample and refinegan
knotty moth
i used 40k sample and refinegan
read the pretrain description, it is 32k
so you should use 32k in the preprocess
azure patio
read the pretrain description, it is 32k
ohhhh that makes sense thennn, thanksss
but what if i dont use applio for conversion? can i use the refine?
and is there much difference using hifi-gan instead of refineGAN?
west bison
AMD Radeon(TM) Vega 8 Graphics
Windows 11, Laptop
im trying to use the realtime voice changer, but the outputting audio keeps getting cut
i used this tutorial https://youtu.be/SxdnGxicJOg?si=9EA6ujmKQbLToCIN
Chunk: 128
Extra: 16384
hallow thistle
AMD Radeon(TM) Vega 8 Graphics
Windows 11, Laptop
im trying to use the realtime ...
This AMD Radeon GPU is an integrated GPU, often found in AMD Ryzen systems. A mobile dedicated GPU is something like AMD Radeon RX xxxxM. Are you sure there is any other GPU in your laptop?
If your laptop has just integrated GPU, it's better to go for online option like Kaggle for superior voice changer performance. Attempting to get any W-Okada DirectML version to work with a integrated GPU would perform a fraction of your AMD Ryzen CPU, or in rare cases CPU-only.
west bison
If your laptop has just integrated GPU, it's better to go for online option like...
Oh alright thanks it seems it's better to try out kaggle
hallow thistle
Oh alright thanks it seems it's better to try out kaggle
west bison
See https://docs.aihub.gg/realtime-voice-changer/cloud/tg-develops-w-okada-fork-...
Damn thanks
tepid sable
I'm sorry, do you know how can we import weights models into applio ?
remote mural
-colab
latent trench
👋
shadow river
What's the best way to generate celebrity AI voices with Play HT? Are there any good GitHub repo with audio files or something to train the models or should I only find YouTube clips to upload? Thanks in advance 🙂
open sun
Why does it seem like every voice sounds terrible even tho the audio from the creator sounds amazing, i have tried almost everything
latent trench
👋
fleet cedar
is vonovox better than w-okada
safe glacier
could someone help me with tg develop w-okada? when i launch the program this error happens
Could not load library with AVX512 support due to:
ModuleNotFoundError("No module named 'faiss.swigfaiss_avx512'")
2025-12-10 17:29:44,181 INFO undefined undefined undefined undefined [loader] Loading faiss with AVX2 support.
2025-12-10 17:29:44,199 INFO undefined undefined undefined undefined [loader] Successfully loaded faiss with AVX2 support.
2025-12-10 17:29:44,203 INFO undefined undefined undefined undefined [init] Failed to load GPU Faiss: name 'GpuIndexIVFFlat' is not defined. Will not load constructor refs for GPU indexes. This is only an error if you're trying to use GPU Faiss.
viral mason
is vonovox better than w-okada
Yes, it's Nvidia only tho
west bison
See https://docs.aihub.gg/realtime-voice-changer/cloud/tg-develops-w-okada-fork-...
Is it possible for it to save my model and settings? Even after I stop running it and start it again
hallow jackal
mellow ermine
-colab
fleet cedar
Yes, it's Nvidia only tho
i have a 5070
supple jacinth
I want to create my own ai cover but I don't know where to start. I prepared mp3 files of the vocals I obtained to train but I am lost on which ai sites to use and how to utilize them without a GPU.
fleet cedar
i need help, vonovox isnt opening for me
i clicked the start.bat
i already ran the setup.bat
nvm, but i tried opening a model and it wont give any audio when i talk
whats the problem
the advanced settings if u prob want it
subtle vapor
hi i installed the thingy for the nvidia model right and i used winrar extracted it then i try open start.bat it says some stuff then i click run the other option is to extract all? and then a black prompt screen pops up for a sec then shuts off
opaque drum
Does anyone have basic models for 100m-200m parameters, for code/Ukrainian language testing? I need a basic model to further train and use for my own purposes
scarlet wedge
why applio's notebook for kaggle is on github and the page doesn't exist?
hallow thistle
why applio's notebook for kaggle is on github and the page doesn't exist?
You sure that's the correct way to access the .ipynb file on GitHub?
scarlet wedge
idk, i just clicked the link in the -kaggle command
rain dagger
Who made it?
hallow thistle
idk, i just clicked the link in the -kaggle command
A Kaggle ipynb file is supposed to be imported within Kaggle website. You wouldn't launch the notebook file directly on GitHub, unless you know what you gonna code inside ipynb file for some implement.
open sun
My voice seems to cut off a little at the end like if i say "Hello guys" it sounds like hello guyz or so and that makes it sound like ai
Using Vonovox btw.
fleet cedar
Using Vonovox btw.
how did u make it give audio bruh
mine doesnt give audio when i talk
fleet cedar
nvm, but i tried opening a model and it wont give any audio when i talk
help
open sun
mine doesnt give audio when i talk
I just selected my input device and output device and press start 😭
fleet cedar
I just selected my input device and output device and press start 😭
same bruh
show ur settings
vonovox settings
fleet cedar
Send a ss of your settings
open sun
You selected a model right?
fleet cedar
ye
open sun
can u show the voice setting if u scroll down
fleet cedar
can u show the voice setting if u scroll down
open sun
Where are you using it in or nowhere for now?
fleet cedar
tf u mean
open sun
as output set your headset or whatever ur using to hear stuff
open sun
u cant hear whatevers going to virtual cable
open sun
if u want to hear yourself then u can use an app thats included in virtual cable, its called audio repeater (MME) and there u select wave in the line 1 and wave out ur headset or whatever
open sun
but how will i talk to ppl with that
put virtual cable as output and select it as input where u wanna talk