all the programs got their guides links up above in the bot message
#✨│ai-help
low shard
viral mason
click whichever (vonovox should work fine, it's the first guide just read up and find the download)
ruby valve
I finally got vonovox downloaded but I am not able to hear anything unless i change back to my normal audio
low shard
I finally got vonovox downloaded but I am not able to hear anything unless i cha...
could u please show ur audio setup? wdym with "change back to my normal audio"?
!give-media-perms 1h @ruby valve
humble beacon
I have click on vonovox guide and it's open a link but were to download
low shard
I have click on vonovox guide and it's open a link but were to download
you need to read the guide, it will give you all the instructions and download links
viral mason
I have click on vonovox guide and it's open a link but were to download
https://github.com/dr87/Vonovox/releases/tag/v1.6.9 go here, but still read the guide
humble beacon
I am new on discord so 😁
viral mason
you need to read the guide, it will give you all the instructions and download l...
btw Nick, do they just need the sc.zip or both files when downloading vonovox
low shard
https://github.com/dr87/Vonovox/releases/tag/v1.6.9 go here, but still read the ...
they need more than that
torn nebula
im highk slow
low shard
they need VAC Lite
torn nebula
i cant figure out how to download vonox
low shard
btw Nick, do they just need the sc.zip or both files when downloading vonovox
only the source code .zip, not tar.gz
low shard
i cant figure out how to download vonox
did you read the guide? you need to firstly download vac lite
torn nebula
did you read the guide? you need to firstly download vac lite
where
torn nebula
did you read the guide? you need to firstly download vac lite
i already have virtual audio cable tho
viral mason
reading is a difficult skill
torn nebula
i cant figure out how to download the thingy itself
humble beacon
😆
viral mason
the guide has the download for both vac lite and vonovox @torn nebula
torn nebula
it shows me how to install redit restributional smt and virtual audio cable but not the voice changer itself i think
torn nebula
the guide has the download for both vac lite and vonovox <@502122059778883598>
thats where im at bro
viral mason
if u need the download link for vonovox here ya go
https://github.com/dr87/Vonovox/archive/refs/tags/v1.6.9.zip
torn nebula
is that not virtual audio cable?
torn nebula
if u need the download link for vonovox here ya go
https://github.com/dr87/Vonov...
thx
do i have to reinstall virtual audio cable
or is vac lite better
ima js get it incase
let me delete the other one
humble beacon
Is it 4.65 gb
low shard
or is vac lite better
vac lite is more suggested than vb audio cable for windows users
viral mason
it's a little more confusing tho,
humble beacon
humble beacon
Yes I am right now in phone I will switch later in pc
humble beacon
So that's it is anymore things I need to run
viral mason
now which one of these do i click?
run setup, then after that is done run start
torn nebula
run setup, then after that is done run start
do i restart pc after?
viral mason
nope
low shard
there isnt any start_https
that was only in original wokada
torn nebula
it feels like it implemented it instantly
low shard
you won't see it again
low shard
i ran it as admin
don't run it as admin
torn nebula
shi too late
viral mason
no after the vac thingy
I don't think you have to do that for vac lite
viral mason
so i did the same to this one
I think u can just run it as normal, do the same file u did but don't run as admin
humble beacon
Bro i have one more prblm i have use many voice changer and my real sounds driver are getting some prblm after deleting all so it's sounds slowly and my mouse click also so any help to get all back and normal
viral mason
if it runs slow u could try switching to wokada deiteris
humble beacon
Can i reset
viral mason
?
humble beacon
No my real pc sounds have get prblm without use any voice changer it's sounds slowly
viral mason
wdym
humble beacon
Do you now how to delete Or reset
All driver
Ok no problem tomorrow I try vonovox in my pc and if I need any more help I msg you to help 😊
tawny onyx
who can help me, why voice changer dont work
viral mason
did u get it off a yt tutorial
ruby valve
viral mason
u got the mic settings kinda wonky it should just be
voice changer:
input: microphone
output: line 1
also set pitch at 0 for all models unless they are female which only work best at ranges from 3-10
12 is usually too high
ruby valve
viral mason
ok so now for the settings
blocksize: 3.0/2.7
crossfade duration: 0.15
extra time: 4.0
I got these from Nick so if anything does not work properly it may be a typo
ruby valve
Ok thanks but now to actually use it lol or hear myself?
viral mason
that idk bc vac lite is weird and u have to go into like your sound settings which is boring so I just use a complicated setup with voicemod, fl studio, voicemeeter and the stuff u already have
voicemod is so I can connect it all with the vb cables and to hear myself
next mist
-colab
patent trellis
torn nebula
@viral mason yo i apologize that i resort to pinging but i have no clue if it works and it doesnt have a monitor setting so i cant hear myself speak
i put it as my headset
the output that is
and i still hear nothing
my input is my mic
it looks totally different and i have no clue where to go
ok got it its now working but is lagging is the cause of that the sample rate?
viral mason
<@1023278814752677918> yo i apologize that i resort to pinging but i have no clu...
U can download voicemod and setup the voice settings like this:
input: line 1
output: headphones
viral mason
Sample rate wouldn't cause that, it's your GPU and settings
hazy current
so likeeeeee,how do i create my own ai voice model?
torn nebula
Sample rate wouldn't cause that, it's your GPU and settings
its not lagging anymore must've been cuz i was streaming to my friend
what does sample rate btw
viral mason
so likeeeeee,how do i create my own ai voice model?
Applio, u can use it on your own PC locally or on a web browser like Kaggle
viral mason
what does sample rate btw
It's just upscaling the model's output to make it sound better
hazy current
Applio, u can use it on your own PC locally or on a web browser like Kaggle
is it appolio.org ?
viral mason
is it appolio.org ?
No, I would ask a moderator as I don't know where to download it locally
torn nebula
No, I would ask a moderator as I don't know where to download it locally
any idea what these do?
hazy current
No, I would ask a moderator as I don't know where to download it locally
man,still hate that we cant use weights for it anymore as they added those lil gem things that i hate
viral mason
man,still hate that we cant use weights for it anymore as they added those lil g...
Weights was never a good option it made crappy models compared to applio or mainline, mainline is kinda lesser known tho as it's older than applio I believe
viral mason
Don't touch those settings they're fine on default, idk what they do anyways
hazy current
Weights was never a good option it made crappy models compared to applio or main...
i just used it cause it was the most user friendly for me
cuz i just had to import some audio files and add some details and then the model would be created
but like
i cant do it anymore
-# mainly cause im too fuckin broke for ts
torn nebula
Don't touch those settings they're fine on default, idk what they do anyways
anything i can change so i improve performance when in games?
viral mason
-# mainly cause im too fuckin broke for ts
Good thing applio is free
hazy current
Good thing applio is free
oh it is?
viral mason
anything i can change so i improve performance when in games?
Gotta lower the game graphics as low as possible until it works well, it works much better in VR tho compared to just running it on a regular flat screen game
torn nebula
Gotta lower the game graphics as low as possible until it works well, it works m...
wait i can run this on VR?
how do i run this on VR?
and doesnt PCVR require like crazy pc specs
viral mason
Just running it as usual and using a VR headset instead of headphones as your mic
viral mason
and doesnt PCVR require like crazy pc specs
Idk I was about to run it on a 1660
torn nebula
i have 2 gpus idk which one is being used tho
on this thing
this is the good one (dont mind CPU usage)
viral mason
Should be your Nvidia since Intel cannot run Vonovox only Nvidia for now
torn nebula
oh alr
viral mason
AMD support will be added eventually
torn nebula
also why is my memory and CPU so high 😭
viral mason
Intel tho basically can't run anything ai
viral mason
also why is my memory and CPU so high 😭
Ai takes a lotttttttttt of power to run locally
That's why when I train models I do it on a web browser so my pc doesn't explode
torn nebula
am i currently training the thingy
i mean i have 16 gb ram
so it should be enough right?
viral mason
The thing from marvel?
torn nebula
The thing from marvel?
no the ai
viral mason
I'm confused
torn nebula
like when i use it
viral mason
Like you're training an ai voice model?
torn nebula
idk you said ai takes a lot of power to run locally so you train models on web
viral mason
Yea
torn nebula
yo forget it im slow fr
viral mason
That's not anything to do with the voice changer tho
viral mason
yo forget it im slow fr
I'm slow too dw
hazy current
so like i booted up appolio but it just shows a blank cmd screen
what do i do :(
anyone?
nvm i was inpatient
so like how do i train a model in applio?
warm iris
I still get the same error. here's some more info:
Command I used(local docker)
docker run -it -p 7860:7860 --platform=linux/amd64 --gpus all \
registry.hf.space/nick088-real-esrgan-pytorch:latest python app.py
I'm using a video with .mov format. The image upscaling works fine, but videos don't.
here are the trace logs:
wintry sphinx
Error creating audio stream: Error opening Stream: Invalid device [PaErrorCode -9996]
Critical error in start_vc: Error opening Stream: Invalid device [PaErrorCode -9996]
Traceback (most recent call last):
File "gui/gui.py", line 1495, in gui.gui.GUI.start_vc
File "gui/gui.py", line 1552, in gui.gui.GUI.initialize_voice_conversion
File "gui/gui.py", line 1606, in gui.gui.GUI.start_stream
File "core/audio/audio_processors.py", line 612, in core.audio.audio_processors.AudioDeviceManager.create_stream
File "core/audio/audio_processors.py", line 602, in core.audio.audio_processors.AudioDeviceManager.create_stream
File "C:\Users\F\Downloads\Vonovox-1.6.9\runtime\Lib\site-packages\sounddevice.py", line 1825, in init
_StreamBase.init(self, kind='duplex', wrap_callback='array',
File "C:\Users\F\Downloads\Vonovox-1.6.9\runtime\Lib\site-packages\sounddevice.py", line 909, in init
_check(_lib.Pa_OpenStream(self._ptr, iparameters, oparameters,
File "C:\Users\F\Downloads\Vonovox-1.6.9\runtime\Lib\site-packages\sounddevice.py", line 2796, in _check
raise PortAudioError(errormsg, err)
sounddevice.PortAudioError: Error opening Stream: Invalid device [PaErrorCode -9996]
wtf does this error code from Vonovox mean i have to do :(((
never tried using ts before smh
tame oracle
i need help with my voice changer
could you please give me some more info?
!howtoask
brittle wing
9060xt 16gb oc windows 11. i was trying to set up the voice changer but after all i couldnt hear myself or anything, https://www.youtube.com/watch?v=SxdnGxicJOg this is the video and no error message just not working
tame oracle
9060xt 16gb oc windows 11. i was trying to set up the voice changer but after al...
thats ur issue, even if the video is only a month old the Voice changer is still over a year old
-rt
tame oracle
you can do either the Tg-develop Fork or deiteris Fork
you just gotta download the AMD version
narrow coyote
can someone help me through apolio please?
how to make an ai voice then use it for singing ai covers
warm iris
I still get the same error. here's some more info:
Command I used(local docker)...
nevermind, I found a fix. I just asked AI what the issue was and it gave me the code to fix it
torpid mountain
hello i want to put VoiceChanger on discord but dont work
hallow thistle
hello i want to put VoiceChanger on discord but dont work
What are your PC GPU and operating system? Did you follow any tutorial/guide before this?
!howtoask
hallow thistle
how to make an ai voice then use it for singing ai covers
-rvc
hallow thistle
There's a website for Applio, and there are also docs about Applio on AI Hub website. https://docs.aihub.gg/rvc/local/applio/
hallow thistle
9060xt 16gb oc windows 11. i was trying to set up the voice changer but after al...
If you have downloaded and run Deiteris/Tg Develop fork W-Okada DirectML, let me know if you have any issue or looking for settings. 
neon quiver
does this thing work on MacBook Pro (13-inch, M1, 2020)
sinful kettle
Hi everyone!
Just an update on my training progress using Mangio RVC locally.
When I used the 5-hour dataset, each training step (epoch) took about 12 minutes and 13 seconds.
After reducing the dataset to around 1 hour and 30 minutes, the training steps dropped to about 2 minutes and 25 seconds each.
Thanks for the support! I’ll keep sharing updates if I face any other issues.
hallow thistle
does this thing work on MacBook Pro (13-inch, M1, 2020)
If you're looking for Applio RVC, there's one that made to run with Mac that has Apple Silicon chip. For W-Okada, both Deiteris and Tg Develop W-Okada forks also work too.
knotty moth
Hi everyone!
Just an update on my training progress using Mangio RVC locally.
Wh...
if only you're okay with the mediocre quality compared to Applio 3.5 with the best pretrain
mangio actually has some bugs and misconfigurations that may affect the model quality results
hallow thistle
Yes, you've seen it right. W-Okada is realtime voice changer and not for training a model, but I didn't said W-Okada could be used to train a model after all. Applio can train voice model.
sinful kettle
if only you're okay with the mediocre quality compared to Applio 3.5 with the be...
I’m just finishing the dubbing process for a Skyrim mod using RVC, and after that, I’ll test Applio.
The community here gave me some feedback about Applio, and I’m quite interested in trying it out.
last lantern
guys im getting crazy because the res on the okada voice changer gets higher and higher and idk what to do
hallow thistle
guys im getting crazy because the res on the okada voice changer gets higher and...
What is your PC GPU? Did you follow any tutorial/guide before this?
last lantern
What is your PC GPU? Did you follow any tutorial/guide before this?
i have a 7600xt and i downloaded the directml version, and yes i followed lots of guides
it used to work fine years ago, but idk why now every voice except the sample ones get such a high res
hallow thistle
i have a 7600xt and i downloaded the directml version, and yes i followed lots o...
Which guide and which version?
last lantern
Which guide and which version?
i have latest version of directml, wait
i tried to do this, but it doesnt work
like i cant run the scripts because it gives me errors
hallow thistle
For better W-Okada DirectML version, try Deiteris or Tg Develop W-Okada fork. https://docs.aihub.gg/realtime-voice-changer/local/deiteris-w-okada-fork/#download-amd-intel-and-cpu-on-windows https://docs.aihub.gg/realtime-voice-changer/local/tg-develops-w-okada-fork/#download-amd-intel-and-cpu-on-windows What you tried to use is the outdated one.
last lantern
Which guide and which version?
i have the v.1.5.3.18a onnxdirectml_cuda
hallow thistle
last lantern
For better W-Okada DirectML version, try Deiteris or Tg Develop W-Okada fork. ht...
oh okay okay, wait lemme see
thank you very much ;D
hallow thistle
Let me know if you have any issue. 
last lantern
thank you a lot i appreciate it ^^
dml-b2332 (click here to download)
i had to click that one, right?
hallow thistle
dml-b2332 (click here to download)
Yes, that one.
last lantern
it downloaded this: voice-changer-windows-amd64-dml
oh okay okay
wait
so i uninstall the other version now?
the one i had before
OOH ITS DOWNLOADING A LOT OF STUFF
yeah this already seems better from the cmd ngl xD
also sorry to bother, but i didnt understand what is the "extra" for
THIS IS SO BETTER BTW THANK YOU SO MUCH! im not having res issues for now, i will try some voices hold on
hallow thistle
So, now I'd have to explain what chunk and extra are for 696th time. 
hallow thistle
THIS IS SO BETTER BTW THANK YOU SO MUCH! im not having res issues for now, i wil...
You're welcome. 
last lantern
So, now I'd have to explain what chunk and extra are for 696th time. <a:oiiacat:...
SORRY
its okay you dont have to
i will just try some stuff xD
just give me a short answer, the higher the extra the better?
hallow thistle
just give me a short answer, the higher the extra the better?
Extra 2.7 s is always recommended; any number above 2.8 s would make the audio to cutting off.
last lantern
thank you very much you're really kind
also, im noticing somethins now
im having no issues with the delay now, but the audio is a bit glitchy even tho my microphone should be quite clean
royal kettle
Hi everyone!
Just an update on my training progress using Mangio RVC locally.
Wh...
Why would you shrink the dataset?
Also 1k epochs is way to many. Max you want is 100.
viral mason
Also 1k epochs is way to many. Max you want is 100.
most of the time my models don't sound good until over 100
near hornet
why applio suddenly cant load?
sinful kettle
Why would you shrink the dataset?
I shrank the dataset mainly to speed up the training time between epochs.
It was the fastest way I found to reduce the time per step.
knotty moth
I shrank the dataset mainly to speed up the training time between epochs.
It was...
no the dataset variation (including pitch range) and quality consistency are more important.
too short or monotonous dataset could lead to less stable or generalization in the results.
knotty moth
Yes, you've seen it right. W-Okada is realtime voice changer and not for trainin...
u prob meant the beatrice trainer that barely ppl manage to use
knotty moth
I shrank the dataset mainly to speed up the training time between epochs.
It was...
you could just use more batch size, though 8 is recommended
woven sail
strange grail
knotty moth
@candid eagle let's move here, does the default pretrain work on you?
humble beacon
Hlo
Bro i have installed vonovox it's working good 👍😊 can you plz tell me best setting for that
Bro help
sinful kettle
you could just use more batch size, though 8 is recommended
Yes, I usually keep the batch size at 8.
With that, my GPU usage goes up to around 98% during training.
That’s why I decided to reduce the dataset instead, to gain some speed between each epoch.
knotty moth
Yes, I usually keep the batch size at 8.
With that, my GPU usage goes up to arou...
unless to remove bad or inconsistent parts, that's not really a valid point
you'd be more likely to sacrifice the model stability (the rate of artifacts/robotic sounds happening)
10 mins is usually bare minimum, or 30 mins/more to be better
late lagoon
hey
can yall help me
with voice changer
whenever im ingame
no one can hear me
i did everything
late lagoon
<@&1159293204038955078>
nocturne mural
why applio suddenly cant load?
halcyon crater
Hi everyone, I have problems downloading the okada voice changer, the command prompt says the version that I downloaded is for removal, which is the version that is currrently working on github? Anyone mind helping me out?
sacred token
hi , someone can help me with the voice changer,
is like the program doesn't recognize my gpu , so only work with the cpu
sacred token
Hi everyone, I have problems downloading the okada voice changer, the command pr...
i downland the version
v.1.5.3.18a.zip
halcyon crater
i downland the version
v.1.5.3.18a.zip
is it working?
but it is not the latest tho, the current latest version is v2 something
sacred token
is it working?
yea
but only had a problem with my graphic card. it does not recognized
halcyon crater
what is your operating system
I heard that there are a lot of versions for all kind of operating system and graphic cards
you might have downloaded the wrong version
halcyon crater
https://www.youtube.com/watch?v=lzb09-j-7qU
refer to this video, it tells you what to download according to your software
viral mason
https://www.youtube.com/watch?v=lzb09-j-7qU
refer to this video, it tells you wh...
Wrong, any yt tutorial is outdated
Uses year old software
gleaming bough
Please Help me find stuff to do please
viral mason
you might have downloaded the wrong version
I can't help as I'm going to work rn but depending on your gpu download one of these
-rt
viral mason
If u use Nvidia 30 series and up use vonovox
If not Nvidia use wokada deiteris as idk anything about tg fork
Anything else is outdated
low shard
https://www.windowscentral.com/microsoft/windows/windows-11-update-killed-localh...
@near hornet the current latest windows version should have fixed that issue, applio works fine on my windows 25h2 locally
gleaming bough
!howtoask
viral mason
The world if people didn't use yt tutorials to get wokada 
gleaming bough
thank you
sinful kettle
unless to remove bad or inconsistent parts, that's not really a valid point
Got it, that makes sense.
I’ve just downloaded and installed Applio, so I’m going to test it now and see how it performs.
marsh garden
I cant extract files
it keep saying The archive is corrupt
i re download for like 5 times already
silver lynx
<@&1159293204038955078>
Full GPU Name: rx 9070 XT 16 GB vram desktop
Operating System Windows 10
Detailed Description: i was trying to use the w-okada fork but for some reason im not hearing anything
viral mason
<@&1159293204038955078>
Full GPU Name: rx 9070 XT 16 GB vram desktop
Operating...
did u get the voice changer from a yt tutorial
west valley
Yurr
west valley
lwky dk-....
viral mason
u can check on ur task manager if u dunno
west valley
aight, ts old so chat pray for me 🙏 😭 (its not old old but it aint new)
viral mason
lemme hear it lol
west valley
Wanna hear how it sounds?
tame rune
Hi, I'm using Applio on Kaggle and I always get the same error after trying to import the GitHub link. After a while, I get an error because I'm not logged in to Ngrok nor do I have the AuthToken (I have an account and the AuthToken). What do I do?: "PyngrokNgrokError: The ngrok process errored on start: authentication failed: Usage of ngrok requires a verified account and authtoken.\n\nSign up for an account: https ://dashboard.ngrok.com/signup\nInstall your authtoken: https ://dashboard.ngrok.com/get-started/your-authtoken\r\n\r\nERR_NGROK_4018\r\n."
west valley
u can check on ur task manager if u dunno
Shi Ionk where ts at— yo we might js gotta thug it out 😭😭😭
viral mason
Shi Ionk where ts at— yo we might js gotta thug it out 😭😭😭
I showed you a video of how to check
west valley
I showed you a video of how to check
hol up gimme a sec
high osprey
Good evening, I'm trying to import my own voice as a model on RVC but I can't do it, I only have my voice in .wav. Do you have a tutorial? I can't find it.
west valley
I showed you a video of how to check
long story short lets js thug it out bru, trust...but if my shi catch fire...at least i tried 😭 
(pray for my pc chat)
thick latch
This option isnt in vonovox docs, what does it do?
viral mason
long story short lets js thug it out bru, trust...but if my shi catch fire...at ...
do u at least know if it's like Nvidia or AMD
west valley
do u at least know if it's like Nvidia or AMD
Nvidia
analog obsidian
This option isnt in vonovox docs, what does it do?
prevents the model from converting noise and random non-voice sounds
basically an ai noise gate
viral mason
west valley
I'll let u download wokada deiteris first and see how good it runs
aight
west valley
https://huggingface.co/Shadicti/deiteris-Fork/blob/main/voice-changer-windows-nv...
alr i got it, what next
west valley
https://huggingface.co/Shadicti/deiteris-Fork/blob/main/voice-changer-windows-nv...
yurr?
west valley
unzip the file and run this file
aight bet
west valley
unzip the file and run this file
its running
viral mason
its running
cool, did it start an app on ur browser?
west valley
cool, did it start an app on ur browser?
Yupppppp
viral mason
alr u need one more thing
west valley
What’s that?
viral mason
alr unzip the file and run that file as administrator (right click and run as administrator)
after that u will have to restart ur pc but u won't have to anymore after that
west valley
alr bet
west valley
after that u will have to restart ur pc but u won't have to anymore after that
Alr i restarted it
viral mason
bet, reopen the voice changer same way u did last time
west valley
It’s open!!!!
viral mason
wait which one?
cable input
spring orchid
-colab
viral mason
aight bet
mk now these depend on gpu but should work good
and all u need to change here is switch to rmvpe
west valley
done chat!!!!!!!!
viral mason
peak
now model downloading and testing time
go look for whatever or whoever u want
https://discord.com/channels/1159260121998827560/1175430844685484042
pls no egirls I will be like a dissapointed father if u choose that
west valley
pls no egirls I will be like a dissapointed father if u choose that
nah im fr boutta pick a carti one ong, don't worry😭
viral mason
np!
west valley
go look for whatever or whoever u want
https://discord.com/channels/115926012199...
alr i found the voice!!!! whats next?
viral mason
alr i found the voice!!!! whats next?
west valley
alr bet!!!!
west valley
last thing...how do i rec?
viral mason
last thing...how do i rec?
?
west valley
like, how do i record vocals to hear how the model sounds?
unreal lintel
Hello, my program doesn't work. i get this error:
[Voice Changer] generate new embedder. (no embedder)
[Voice Changer] use torch contentvec [ONNXRuntimeError] : 7 : INVALID_PROTOBUF : Load model from pretrain/content_vec_500.onnx failed:Protobuf parsing failed.
[Voice Changer] exception! loading embedder PytorchStreamReader failed reading zip archive: failed finding central directory cpu
Traceback (most recent call last):
File "voice_changer\RVC\embedder\EmbedderManager.py", line 44, in loadEmbedder
File "voice_changer\RVC\embedder\OnnxContentvec.py", line 17, in loadModel
File "onnxruntime\capi\onnxruntime_inference_collection.py", line 383, in init
File "onnxruntime\capi\onnxruntime_inference_collection.py", line 424, in _create_inference_session
onnxruntime.capi.onnxruntime_pybind11_state.InvalidProtobuf: [ONNXRuntimeError] : 7 : INVALID_PROTOBUF : Load model from pretrain/content_vec_500.onnx failed:Protobuf parsing failed.
During handling of the above exception, another exception occurred:
Traceback (most recent call last):
File "voice_changer\RVC\pipeline\PipelineGenerator.py", line 30, in createPipeline
File "voice_changer\RVC\embedder\EmbedderManager.py", line 25, in getEmbedder
File "voice_changer\RVC\embedder\EmbedderManager.py", line 48, in loadEmbedder
File "voice_changer\RVC\embedder\FairseqHubert.py", line 11, in loadModel
File "fairseq\checkpoint_utils.py", line 425, in load_model_ensemble_and_task
state = load_checkpoint_to_cpu(filename, arg_overrides)
File "fairseq\checkpoint_utils.py", line 315, in load_checkpoint_to_cpu
state = torch.load(f, map_location=torch.device("cpu"))
File "torch\serialization.py", line 797, in load
with _open_zipfile_reader(opened_file) as opened_zipfile:
File "torch\serialization.py", line 283, in init
super().init(torch._C.PyTorchFileReader(name_or_buffer))
RuntimeError: PytorchStreamReader failed reading zip archive: failed finding central directory
[VCClient] wait web server... done 200
viral mason
like, how do i record vocals to hear how the model sounds?
u just talk into it
unreal lintel
[Voice Changer] generate new embedder. (no embedder)
[Voice Changer] use torch c...
My pc is a laptop, AMD RYZEN 5000 series CPU with amd radeon integrated graphics so i don't have a GPU
and i'm using MMVCServerSIO_win_onnxdirectML-cuda_v.1.5.3.18a
west valley
u just talk into it
Do I need to click a button?
viral mason
yea just hit the start button
west valley
Alr bet thanks bru, you the goat!!!
west valley
yea just hit the start button
i clicked start and it don't look like nun changed, idk if im supposed to see like a file or anything but uhm..yuh-
viral mason
i clicked start and it don't look like nun changed, idk if im supposed to see li...
you're supposed to be able to hear yourself
west valley
you're supposed to be able to hear yourself
i don't for some reason
viral mason
i don't for some reason
try changing this to ur headset/headphones
west valley
try changing this to ur headset/headphones
alr
viral mason
see if that works
lost ember
any1 know the best girl voice model n settings?
west valley
see if that works
am i supposed to be hearing myself while im recording or both non recording and recording?
lost ember
the voice barely works for me
like doesnt even form the words etc
i need a good girl model and settings
unreal lintel
Anyone who could give some tips/knowledge? https://discord.com/channels/1159260121998827560/1431083533027246213
viral mason
am i supposed to be hearing myself while im recording or both non recording and ...
only should hear the voice when it's started and should be the ai only
west valley
only should hear the voice when it's started and should be the ai only
these are all my outputs
AHHH WAIT I GOT IT NOW
neon quiver
If you're looking for Applio RVC, there's one that made to run with Mac that has...
hey whats applio rvc
dull pawn
Hello, I want to create my first model on my phone, but I don't know how.
inland stag
im getting an error whyle installing a voice
failed to import voice because toml was not found
hallow thistle
hey whats applio rvc
Applio is basically RVC WebUI fork, if you didn't know.
-rvc
hallow thistle
Hello, I want to create my first model on my phone, but I don't know how.
It's not quite possible to train an AI model locally on a smartphone, but the online cloud one is sure possible. What model are you looking for? RVC or Stable Diffusion?
hallow thistle
im getting an error whyle installing a voice
!howtoask
hallow thistle
these are all my outputs
I expect "Line 1 (Virtual Audio Cable)" to appear in output, but I only see "Cable In 16ch (VB-Audio Virtual Cable)". 
Has the team VB-Cable fixed their program to be usable with W-Okada or something? Because on the guide it still says Virtual Audio Cable, which is a whole different program.
sonic mauve
.help
brazen gorgeBOT
.fmbot help & command overview
sonic mauve
Anyone here know how to start making Al images? Online or local? Would love some help!
neon quiver
Applio is basically RVC WebUI fork, if you didn't know.
@hallow thistle srry i kinda forgot. are there any tutorials anywhere on how to use this fork thing
hallow thistle
<@561807329977434122> srry i kinda forgot. are there any tutorials anywhere on h...
You wouldn't wanna expect the tutorial from YouTube to be your main guide, because not only it quite rare for someone who has Mac to test certain AI programs locally, some of which can be outdated despite their upload dates.
neon quiver
You wouldn't wanna expect the tutorial from YouTube to be your main guide, becau...
so u guys got tutorials here or
hallow thistle
Of course, there are two way to run either W-Okada or Applio; one is to run the program locally or use a cloud service, which can be faster than your local machine.
This is the "guide" on how to install Applio on Mac. https://docs.aihub.gg/rvc/local/applio/#linux--macos I don't use Apple Mac, so I'm not sure if it works.
When I see simple questions, I still answer with a bit long response. 
stuck saddle
My program is very laggy and i can't seem to get it fixed, i have a Radeon RX 6600 GPU, i've tried changing settings and other things but i just can't get it to sound clear without stutters and be fast to.
knotty hearth
whenever i speak i hear my usual self and not the voicechanger
hallow thistle
Vonovox is not working good
If Vonovox doesn't work on your PC, Deiteris and Tg Develop W-Okada forks are the only reliable ones left. What are your PC GPU and operating system?
hallow thistle
why cant my voicchanger work
!howtoask
hallow thistle
Make sure to read help guidelines before start asking. 
hallow thistle
My program is very laggy and i can't seem to get it fixed, i have a Radeon RX 66...
Did you follow any tutorial or guide before this? And which W-Okada fork/version did you try to run?
humble beacon
i5 12th gen processor rtx2050
Vonovox when I upload model and start then it's work but cut sound and some times does not pronounce the word which I have give
@hallow thistle any solution
hallow thistle
<@561807329977434122> any solution
Use reply feature instead of pinging in separate message.
humble beacon
?
hallow thistle
Have you tried setting Block Size, Extra Time and Crossfade on Vonovox? Because almost everyone who uses Vonovox would left settings to be default.
humble beacon
Yes I try but same problem I think when I speak then vonovox can't cach my voice properly
This same thing work in every rvc voice changer
hallow thistle
I don't use Vonovox so I can't tell if the issue has to be your microphone, your GPU was slow under work or you still didn't set something right. You can try Deiteris or Tg Develops W-Okada fork. https://docs.aihub.gg/realtime-voice-changer/local/deiteris-w-okada-fork/#download-nvidia-on-windows
RVC doesn't always mean realtime voice changer.
If you say the issue "happens" on every other W-Okada, it means you have tried all the original W-Okada (v.1.5.3.18a), Deiteris or Tg Develop W-Okada forks, and even Vonovox. Also, the "RVC voice changer" could be mistaken easily for an old realtime voice changer that came within mainline RVC WebUI from 2023.
hallow thistle
What even is that question?
humble beacon
Give me the best RBC that you use and that works in my PC properly without any mistake and latest version direct download link
🙏
hallow thistle
Give me the best RBC that you use and that works in my PC properly without any m...
I already gave you the link, you might have not seen me did it. https://huggingface.co/Shadicti/deiteris-Fork/blob/main/voice-changer-windows-nvidia-b2332.zip
The first link is the guide for Deiteris W-Okada, while the second and latest link is the download link for the same program, if you still have no idea what I'm talking about.
The program, however, it depends on how you set it to work, the hardware your PC has, so you can't expect it to 100% work flawlessly.
humble beacon
hallow thistle
Also, Tg Develop has the most recent version of fork W-Okada since it's forked from Deiteris one; its UI is different but all functionally the same. https://docs.aihub.gg/realtime-voice-changer/local/tg-develops-w-okada-fork/ https://github.com/tg-develop/voice-changer/releases
hallow thistle
Why would you use this old realtime voice changer that came within mainline RVC WebUI from 2023? This one is old, what the hell.
Inside Deiteris or Tg Develop W-Okada folder is supposed to look like this or this.
"The realtime mode" in mainline RVC and W-Okada are two different programs.
thick latch
lmao
humble beacon
Give me the new version of RVC direct download link
hallow thistle
Nah, you weren't listen to me after all.
If you keep looking for "the realtime mode in mainline RVC" which is the one in your screenshot photo you sent, I won't be helping you about that. I can help about settings in W-Okada and Vonovox. 
humble beacon
hallow thistle
<@561807329977434122> if yes then which one I download
Here, I circled for you. You download these two files. After they downloaded to your drive, you use WinRAR or 7-Zip to extract one of these file, extract to somewhere like "D:/MMVCServerSIO" or "C:/Users/your username/Downloads/MMVCServerSIO".
You still have no idea if "voice changer" means realtime voice changer, right? The name literally says it so.
humble beacon
Oh set I have download 1.36 gb one is that fine
hallow thistle
Don't ask too many questions.
If I say to download two files, it means you should download two files and not only one of these.
humble beacon
Ooh ok and then after then
hallow thistle
I won't be repeating the same step for the second or third time. If I do so, it means you didn't listen.
humble beacon
Ok wait I say I have to download that too file and then extract then what to do
Sorry for not understanding
hallow thistle
If these two files are now in your download folder, you extract one of these zips with WinRAR or 7-Zip, and the program will combine these two zips into one single zip.
humble beacon
Can't open
hallow thistle
humble beacon
Ok i got it 👍 thanks
hallow thistle
When you double click on MMVCServerSIO to run the program, there are a few downloads within the terminal window at the first time you run the program. After they finished downloading, the program will open your browser, which is completely normal for this specific W-Okada realtime voice changer.
On Tg Develop W-Okada, the UI looks like this. https://cdn.discordapp.com/attachments/1426425409498775552/1426490804330954762/image.png?ex=68fc8e16&is=68fb3c96&hm=a4f2035bae86d336ac9b36476be0cf9ec0d1ab527eb0bc8ae5f00357f1c63f70&
Here's your settings:
Chunk: around 110 - 130 ms
Extra: 2.7 s
GPU: NVIDIA GeForce RTX 2050
F0: rmvpe
Input Device: your microphone
Output Device: Line 1 (Virtual Audio Cable)
Monitor Device: optional, you can set this to your main speaker/headphone to hear W-Okada.
humble beacon
Fail to download
hallow thistle
Fail to download
Try these solutions from this guide. https://docs.aihub.gg/realtime-voice-changer/local/tg-develops-w-okada-fork/#how-to-fix-failed-to-download-or-verify
queen arch
hey guys im tryna make a cover of a greenday song with a billie joe armstrong voice and everytime a harmony comes in it freaks out
anyone know for help with this
hallow thistle
hey guys im tryna make a cover of a greenday song with a billie joe armstrong vo...
Weights website or Applio RVC? Weights has its own main vocals/chorus vocals separation; it also has reverb remover. But the Applio RVC one needs another program, UVR5, to manually extract vocals, instrumental or main/chorus vocals from an audio.
queen arch
Ahh im new to this but i believe applio RVC
would u be able to send UVR5 in chat and weights?
hallow thistle
would u be able to send UVR5 in chat and weights?
How do you run Applio RVC? Local or online?
queen arch
How do you run Applio RVC? Local or online?
uhh
i watched a youtube video
and extracted this rvc file
it’s like a little app
i added in my model
and the vocals i wanted
but i listened to them on media player and it sounded good until the harmonies came in
and it like glitched it out alot
went 3 octaves up and voicecracks haha
hallow thistle
Running an AI program on your PC means locally, while running on an online notebook means cloud service.
queen arch
nvidia rtx 3060ti
queen arch
Your GPU is good enough to run Applio or UVR5 locally.
thanks!
would you also be able to send the weights thing pls
and another quick question i had
hallow thistle
queen arch
i have a sound file with which id like to train my own model
i’ve tried to find how but i can’t seem to find a method
i chopped up master tracked vocals clean and put it as about a 13 minute clip of pure album vocals
hallow thistle
The thing is that, you'll have to subscribe to Weights service to have all extracted vocals, main and chorus vocals, and instrumental tracks be downloadable from your account.
queen arch
training M’s S’s and stuff
queen arch
The thing is that, you'll have to subscribe to Weights service to have all extra...
ah
im broke going for free route hahaha
but anyway to train my own ai pls?
hallow thistle
I have never trained voice model, so I'm not sure what to do about this kind of topic, besides naming the model name and set a path to dataset folder in Applio. For training, you can try ask some other staff members or fellow model makers if they know how to make a voice model to sound good.
queen arch
thank you for your help!
dull pawn
I think you can easily create voice models on the phone.
queen arch
I have never trained voice model, so I'm not sure what to do about this kind of ...
also btw how do i do the search voice thing?
in find voiced
voices*
viral mason
also btw how do i do the search voice thing?
https://discord.com/channels/1159260121998827560/1175430844685484042 use the search bar
barren meteor
hello. i was wondering if its possible to train a better voice model yourself than the publically available ones. is there someone who has made a better voice model than the ones available in weights dot com?
viral mason
hello. i was wondering if its possible to train a better voice model yourself th...
you can check here for good voice models
https://discord.com/channels/1159260121998827560/1175430844685484042
I also create very good voice models which u can see all of mine on weights.gg
https://www.weights.com/user/your_local_worm
viral mason
but if u wanna make sure they're high quality ones of a character that isn't available u shouldf make one on ur own
queen arch
i’ve got like 14 mins of pure vocals i really wanna train it
viral mason
how do i make my own voice?
Applio
viral mason
is it free
yes
viral mason
it's what most good model creators use
viral mason
And let it run?
it's not as simple as what weights has bc weights is ass
queen arch
i know haha
viral mason
At least for making models
queen arch
it's not as simple as what weights has bc weights is ass
could u send me link pls
and possibly a video tutorial?
thanks man
viral mason
but go ahead and make an account on kaggle as it's the easiest way to learn training with it
https://www.kaggle.com
and you'll need an account with https://dashboard.ngrok.com
queen arch
what are those for?
viral mason
kaggle is what u will use applio on just for now since that's how I know how to use it and once u learn it u can ask others if they know how to run it locally on pc
viral mason
and ngrok is just a free token you'll need for running the training part
queen arch
is there a video showing how to set it up?
viral mason
could you send the main applio link pls also
there is no main link or tutorials but I could record some bits of how to use it
this is how to import it after making an account
on kaggle btw go to settings and phone verify or u can't do anything
viral mason
I don't know the link for it
queen arch
ah okay all good
thick latch
The only thing which affects MS/response time the block size?
true heron
hey is there a way to make ai voice sound less robotic?
pastel moth
Ive downloaded wan2.2 rapid also their example.json workflow but cant seem to select checkpoints. Also the safetensors doesnt appear in the checkpoint list in the model library
tame rune
Help, I tried to access the Google Collab app and I got this error: "Fetch for https ://api.github.com/repos/iahispano/applio/contents/assets?per_page=100&ref=master failed: {
"message": "No commit found for the ref master",
"documentation_url": "https ://docs.github.com/v3/repos/contents/",
"status": "404"
}
CustomError: Fetch for https ://api.github.com/repos/iahispano/applio/contents/assets?per_page=100&ref=master failed: {
"message": "No commit found for the ref master",
"documentation_url": "https ://docs.github.com/v3/repos/contents/",
"status": "404"
}
at new NO (https ://ssl.gstatic.com/colaboratory-static/common/a2ed59bb479fa4e50cbf6b88cfed4d6a/external_binary.js:2678:68)
at sa.program_ (https ://ssl.gstatic.com/colaboratory-static/common/a2ed59bb479fa4e50cbf6b88cfed4d6a/external_binary.js:2661:436)
at ta (https ://ssl.gstatic.com/colaboratory-static/common/a2ed59bb479fa4e50cbf6b88cfed4d6a/external_binary.js:15:57)
at sa.next_ (https ://ssl.gstatic.com/colaboratory-static/common/a2ed59bb479fa4e50cbf6b88cfed4d6a/external_binary.js:13:196)
at raa.next (https ://ssl.gstatic.com/colaboratory-static/common/a2ed59bb479fa4e50cbf6b88cfed4d6a/external_binary.js:16:48)
at b (https ://ssl.gstatic.com/colaboratory-static/common/a2ed59bb479fa4e50cbf6b88cfed4d6a/external_binary.js:16:309)"
please elaborate more your pc gpu, operating system and the google colab jupyter notebook link you're using
low shard
<@132858642310365184>
Please don't ping random admins
low shard
Who can help with rvc Voice Changer????
!howtoask
tame rune
please elaborate more your pc gpu, operating system and the google colab jupyter...
windows 11, 8 GB ram, Intel(R) Celeron(R) N4020 CPU @ 1.10GHz 1.10 GHz. But I don't think it's a problem with my GPU. Yesterday I was able to train a model on the same page, but now I get this error and I can't access the colab
worldly locust
hey question. does anyone know how to use weights models with MMVC? I was so used to appilo before
elder cloak
okay so idk why I keep getting an error on RVC AI Cover Maker UI. Can someone help?
dull pawn
Hello, I want to create a voice model for a singer, but I don't know how to create one. I only have one phone. Please help me.
queen arch
Weights website or Applio RVC? Weights has its own main vocals/chorus vocals sep...
hey so i got uvr5
it lets me isolate vocals
but i cant seperate the harmonies
which is what my problem is
all it did was let me remove vocals from a song which i can already do
crimson depot
Can I run cover artists locally on my pc with a 1660ti?
latent kettle
Can I run cover artists locally on my pc with a 1660ti?
You mean what ?
hallow thistle
You mean what ?
He means to make RVC AI cover locally on his PC.
hallow thistle
Can I run cover artists locally on my pc with a 1660ti?
Yeah, sure, Applio RVC can work with GeForce GTX 16 GPU.
-rvc
queen arch
<:karinthink:1015964749470650400>
?
hallow thistle
Sorry, but asking me with just "?" doesn't make me think of any proper answer.
hallow thistle
I'm not sure how the other guy helped you after me. He just told you to go for Kaggle for whatever reason. There are a lot of UVR5 models you can use in UVR5; some of which have their own algorithm and how they separate into stems (vocals/instrumental, main/background vocals, vocals/background noise, or even bass/drums/other/vocals). https://docs.aihub.gg/rvc/resources/dataset-isolation/#local-uvr
crimson depot
Yeah, sure, Applio RVC can work with GeForce GTX 16 GPU.
How do I put the collab files and run it on my pc?
knotty moth
Sorry, but asking me with just "?" doesn't make me think of any proper answer.
that also sometimes made me irritated for some reason like if they were saying "idfk what you're talking about", though it's not really in this case
crimson depot
I want to cover on my pc without collab
How can I use those files on my PC offline?
hallow thistle
Mate, slow down. I'm trying to find a proper answer.
knotty moth
I want to cover on my pc without collab
I forgor the link to the cover maker, but your 1660 Ti should be capable enough
crimson depot
I have Python, ffmpeg, etc
hallow thistle
To use Applio in your PC, download Applio from Hugging Face. https://huggingface.co/IAHispano/Applio/tree/main/Compiled Follow steps from the guide. https://docs.aihub.gg/rvc/local/applio/#download--installation
This is Applio.
knotty moth
I want to cover on my pc without collab
this is the link to the latest release of Applio: https://github.com/IAHispano/Applio/releases/tag/3.5.0
hallow thistle
After finished downloading the zip file, extract the zip file with WinRAR or 7-Zip to somewhere like D:/Applio or C:/Users/username/Downloads/Applio.
In Applio folder, double click on "run-applio.bat" to launch the program and open your browser. Do not run as administrator.
crimson depot
It doesn't seem very complicated. Does that need to download something or does it already come in the compressed file?
crimson depot
I use a stems separator with that same architecture and when I'm going to use a model you have to download it. That's why my question of whether this discharges something the first time or something like that
hallow thistle
I don't understand about your statement. 
crimson depot
I'm using a translator
hallow thistle
crimson depot
Nice
elder cloak
🙁
hallow thistle
To upload a voice model into Applio, you can go to Download tab, there are two options on how you upload the voice model.
crimson depot
My internet is exceedingly bad.
crimson depot
Cuba
elder cloak
okay so idk why I keep getting an error on RVC AI Cover Maker UI. Can someone he...
@hallow thistle need help
hallow thistle
<@561807329977434122> need help
There's another RVC fork named Applio, if you haven't tried it yet.
!howtoask
elder cloak
@hallow thistleoh okay thanks i will! 🫡
hallow thistle
Hello, I want to create a voice model for a singer, but I don't know how to crea...
You have been asking the same question for like 5 times in different channels. 
You also tried to tell someone it was possible to train a voice model on a phone, yet your question, which is an irony. 
queen arch
there is no main link or tutorials but I could record some bits of how to use it
didnt work lol
and my thing is on lightmode which is just horrible lmao
queen arch
You have been asking the same question for like 5 times in different channels. <...
alot of the staff in here dont really know how to fix my problem at all
ig hes just asking around idk
hallow thistle
alot of the staff in here dont really know how to fix my problem at all
The thing is I once responded to him, and he didn't answer anything back.
viral mason
and my thing is on lightmode which is just horrible lmao
Turn on darkmode in the settings
queen arch
Turn on darkmode in the settings
still doesnt work for me
theres some things in the video u have in kaggle but didnt show how
what do i do now
knotty moth
if you need step-by-step guidance, I'd recommend to post a thread in https://discord.com/channels/1159260121998827560/1192011222023950368
hallow thistle
To start a Kaggle notebook session, navigate to "Run" menu and then click "Start session.
queen arch
To start a Kaggle notebook session, navigate to "Run" menu and then click "Start...
i’ve done that
but thanks
idk why it won’t work
empty jasper
whats the best girl voice model
barren meteor
hello. i've decided to train a model and to gather the dataset i went over to youtube. the guides say that a wav or flac format is preferred but conversion to them doesn't restory audio quality. is there any way that i can download the audio i want from youtube in wav or flac format? or any other solution possibly? thank you
potent meadow
can somone send the link to do vocal transforming ai because cant find it
hallow thistle
hello. i've decided to train a model and to gather the dataset i went over to yo...
Audio of a video downloaded from YouTube mostly encoded as aac or opus, which are both lossy audio codecs. Attempting to download an audio file as wav/flac would have that tool or service to simply convert from aac or m4a into wav/flac, which doesn't fully restore its original audio quality but also unnecessarily increases file size in the end. The aac/m4a one is the highest audio quality you can get from YouTube.
The only way to format raw audio into a proper audio dataset possible is to manually edit them using Audacity or another audio program available, and then export it into wav. Additionally, if that one audio has background instrumental, noise or unwanted elements, use UVR5 with specific UVR models to extract audio into stems. https://docs.aihub.gg/rvc/resources/dataset-isolation/
hallow thistle
can somone send the link to do vocal transforming ai because cant find it
-colab
hallow thistle
whats the best girl voice model
jade storm
Hello, dumb and simple question here. What's a recommended realtime voice changer to use..? I'm new to all of this and I see a few online.
(I'm sorry if I'm in the wrong chat asking this)
hallow thistle
Hello, dumb and simple question here. What's a recommended realtime voice change...
Deiteris and Tg Develop W-Okada forks.
!howtoask
stable quartz
Hey sorry 2 ask but my friend and i wanna use tts to make a family guy video how do i use the tts on weights
As i cant seem to find it
Im trying to find tts ai that can convey emotion in voice in a way and make sounds such as heavy breathing scream cry cough
low shard
windows 11, 8 GB ram, Intel(R) Celeron(R) N4020 CPU @ 1.10GHz 1.10 GHz. But I ...
That's a CPU, not a GPU, but considering how old that CPU is I'm guessing you got no dedicated GPU to run it locally, so yeah cloud is your only way
I just tried running the Applio UI Colab, it seems all fine
and btw, i'd suggest the Applio UI Kaggle or Lightning.AI for more free gpu time for RVC Training
low shard
hey question. does anyone know how to use weights models with MMVC? I was so use...
Could you please show a screenshot of your program? MMVC Could mean a different version of wokada, there are many different ones, also I'm guessing you meant to say Applio
low shard
help my ping is high
!howtoask
unique dew
hi, I cant open my rvc, it stucked. What should i do?. I'm new.
viral mason
Is vonovox better?
it's currently the best ai voice changer for Nvidia yes
viral mason
Tysm, will use and try that, used the other one yeaaaars ago
if u have any questions just ask here or u can dm me ^^
lavish ivy
anyone knows why the voice changer fully shuts down? every time i boot it through start_http it pulls up cmd and shuts itself off after a second or two
viral mason
anyone knows why the voice changer fully shuts down? every time i boot it throug...
Because you're using a outdated version 💥
What gpu do u have
lavish ivy
What gpu do u have
i have a 7800xt, i just followed the one in #1393389200862089240 hoping that it'd work 
viral mason
i have a 7800xt, i just followed the one in <#1393389200862089240> hoping that i...
what is an xt?
is that a laptop or smth, I really only know of Nvidia, AMD, and Intel when it comes to gpus
lavish ivy
what is an xt?
the xt is a model just above the normal 7800 from amd, it's like any other desktop gpu
viral mason
the xt is a model just above the normal 7800 from amd, it's like any other deskt...
so it's amd? sorry if I'm slow lol
if it is u can use wokada deiteris fork, the third guide
-rt
viral mason
applio on kaggle gives me the same link for all 3 so I can only open either 3 applios, 3 tensorboard, or worst optiom, 3 file browsers
@low shard whenever you're available could u look into this issue to see if it also happens for u?
lavish ivy
if it is u can use wokada deiteris fork, the third guide
oh this one i have tried as well, it boots fine but there is not audio coming out and im not sure of it registers my input
viral mason
oh this one i have tried as well, it boots fine but there is not audio coming ou...
did u have your mic settings setup correctly?
lavish ivy
did u have your mic settings setup correctly?
my input is set to my focusrite, and the output is set a cable input from vac
i tried monitoring through the focusrite too but nothing comes through
viral mason
my input is set to my focusrite, and the output is set a cable input from vac
i...
you could try downloading voicemod and turning on the "hear myself" option
lavish ivy
you could try downloading voicemod and turning on the "hear myself" option
voicemod works perfectly fine with both my focusrite and even routing through vac
grand stag
I saw a tutorial for RVC model training that said to manually split your audio file used for training into clips of 10 seconds or less and to avoid cutting in the middle of a spoken word. My understanding is that the WebUI for RVC has a process that automatically splits the training audio file into about 4 second clips. So, is manually splitting the clips into 10 second clips beforehand just unnecessary?
low shard
applio on kaggle gives me the same link for all 3 so I can only open either 3 ap...
this issue isn't happening to me, could you retry?
low shard
applio on kaggle gives me the same link for all 3 so I can only open either 3 ap...
@nocturne mural you might want to check if the issue is happening to you
viral mason
this issue isn't happening to me, could you retry?
I tried deleting my account and making a new one and the issue followed me still 💔
knotty edge
I tried deleting my account and making a new one and the issue followed me still...
hey again how do i install w okada fork again i did a lot of changes and a lot of cleanups and stuff
nocturne mural
<@989772388508000306> you might want to check if the issue is happening to you <...
I'll try to check it out.
viral mason
hey again how do i install w okada fork again i did a lot of changes and a lot...
Heloo, what gpu do u have again? I don't remember
vapid gust
could someone help meo ut
whenever itruncate the silence, the tracks do not align together
nocturne mural
this issue isn't happening to me, could you retry?
Even by executing the cell several times, I cannot replicate the "error".
vapid gust
help
nocturne mural
@viral mason Are you sure you only modified the value of ngrok_authtoken with your token and didn’t change the entire cell?
viral mason
<@1023278814752677918> Are you sure you only modified the value of ngrok_authtok...
I changed nothing but adding my token
viral mason
maybe it just does not like me
low shard
maybe it just does not like me
delete the notebook and retry from scratch
warped elbow
what is the best girl voice changer plz and settings
queen relic
any sora invite code
mild heron
start
misty turret
chat why the voice changer got hella lag like every time i try to talk it sounds really laggy
how can i fix this
hexed gazelle
i have the same problem to I have everything setup correctly and my Res is at like 20k ms
sterile yacht
Do someone who uses Okada Voice Changer (vcclient_win_cuda_2.1.4-alpha) knows how to use the GPU instead of CPU when running the models? My gpu don't is not being showed as an option
Meanwhile in the old versions it is being showed
sterile yacht
this is ancient tech, outdated as hell
Really? I downloaded this thinking it was the latest version
viral mason
yea the code is extremely old, there's 3 new voice changers those being wokada deiteris, wokada tg fork, and vonovox
sterile yacht
yea the code is extremely old, there's 3 new voice changers those being wokada d...
What is their differences if i may ask
finite badge
hi can anyone help me download the voice changer
inner surge
the voicechanger dont work for me only passtrought
i did all steps in a vid
any help ?
turbid fractal
Does anyone know of a version of rvc that supports amd GPU's, I had one before that used a web gui
Specifically it worked with windows
analog zephyr
anyone got a k9 voice changer?
inland flame
Can someone help me so-vits. I installed the current 2025 verison though a bat file. And when it was done installing? I had to find it thought the search menu on windows 10.
But when i launch it. It opens a cmd tab. And runs a bunch of code i cant read before closing
viral mason
What is their differences if i may ask
My bad I got super busy, Vonovox is currently the only one of the three currently still receiving updates and has the most advanced features of all of them but only has Nvidia support, Wokada tg fork has a little extra stuff added from the wokada deiteris base, and wokada deiteris is the original wokada if it was actually stable and worked good
dapper mason
does the voice changer stuff work on mac?
tired merlin
what should i make my settings
shadow spade
does the voice changer stuff work on mac?
it should
knotty moth
Really? I downloaded this thinking it was the latest version
it was the "latest version" 3 years ago, not now
glossy hull
sora code??
viscid tangle
hi can someone help me everytime i open the bat file it just closes
simple ore
AMD Radeon (Vega 8)
you need DML version
simple ore
I think there's some issue with repository cloning rn
oak edge
I think there's some issue with repository cloning rn
😭 ayo what
oak edge
I think there's some issue with repository cloning rn
like what does that mean?
simple ore
like what does that mean?
did you even run the install cell?
vapid gust
@simple ore do u know tensorboards
simple ore
where is overtraining
tensorboards do not really show 'overtraining'
use and grads 50
to see weirdness
oak edge
did you even run the install cell?
YES
vapid gust
ive been told to check loss/g/total to check for plataeus and then if it rises thats overtrainijg
oak edge
@simple ore
simple ore
anything weird?
how big is the dataset, how often did you save it?
vapid gust
its a 20 minute studio recorded vocal dataset
simple ore
how big is the batch size / how many steps per epoch?
vapid gust
batch size is 8
steps per epoch is 15
its roughly at 2110 epochs right now
steps 36057
simple ore
its roughly at 2110 epochs right now
vapid gust
lMFAOO
oak edge
oak edge
something buggy with your code it seems
vapid gust
i left it over night and i fell asleep 😭
simple ore
<@155030383648440320>
hold on
oak edge
smt wrong w my code?
no no with the original code for kaggle for tunnel
vapid gust
hows it wrong?
oak edge
hows it wrong?
it ain't working for me, so i had to go and cry to chat gpt to give me a working code
vapid gust
it worked for me
oak edge
and the code gpt gives works perfectly
simple ore
stop using chatgpt for code, for fuk sake
simple ore
one idiot asking another
simple ore
did you save using default every 10 epoch?
oak edge
stop using chatgpt for code, for fuk sake
😭
simple ore
if yes, throw away anything over 200e
vapid gust
simple ore
I did not say what you can save
you ran training
during training the default is 'save a model every 10 epoch'
get all those saves
hopefully you also saved the actual voice model every 10 epoch
so test what you got
vapid gust
during training the default is 'save a model every 10 epoch'
mines set at 15
and no i didnt actually download the model on to my system
simple ore
did you train on kaggle/colab?
vapid gust
kaggle
simple ore
well, download what you have
simple ore
like what does that mean?
I've imported the latest applio kaggle notebook and it works fine
oak edge
I've imported the latest applio kaggle notebook and it works fine
then what could be going wrong with mine, i only had to resort to gpt cuz the first one bugged up
vapid gust
@simple ore where do i the files save to when i press uplaoad?
simple ore
then what could be going wrong with mine, i only had to resort to gpt cuz the fi...
dunno, open kaggle, create new notebook, then import notebook
oak edge
dunno, open kaggle, create new notebook, then import notebook
what about my 8.3gb data
i worked 1 week on creating
oak edge
well, if you've enabled persistence, then you need to read what it does
i mean i got variables and persistence on but how do i work with it
simple ore
Vidal had an explanation why that error happens
simple ore
somewhere here #✨│ai-help message
oak edge
hey gng
viscid tangle
hi my virtual cable is not connecting to discord
warped elbow
whats the last version of voice changer client demo please
nocturne mural
hey gng
Could you send a screenshot of what you have inside /kaggle/working, and then, if the folder /program_ml/ exists, expand it and show its contents as well?
oak edge
Could you send a screenshot of what you have inside /kaggle/working, and then, i...
sure one sec
oak edge
Now you know why app.py doesn't exist.
so it's not getting installed?
oak edge
Now you know why app.py doesn't exist.
like this is the stuff that's there after i click in install column
nocturne mural
so it's not getting installed?
in the first place, you shouldn’t have run program_ml more than once, since that caused it to duplicate. if you look closely at the screenshot, you’ll see there’s another program_ml.
oak edge
in the first place, you shouldn’t have run program_ml more than once, since that...
OH MY GOSH YES
AND There is the app.py inside it
nocturne mural
try to see how many copies you’ve created, and whether you’ve trained your models in those copies or not.
oak edge
try to see how many copies you’ve created, and whether you’ve trained your model...
i think only one copy
nocturne mural
well, then if you don’t have anything in that copy, just create a cell and run this:
!rm -r /kaggle/working/program_ml/program_ml/```that will move all the files to program_ml and delete the copy, which means you’ll get your app.py back, as well as restore all the files (in case you deleted any).
so now you need to revert the code in the 'start applio' cell back to its original version, not your ChatGPT version.
oak edge
well, then if you don’t have anything in that copy, just create a cell and run t...
okiee
oak edge
well, then if you don’t have anything in that copy, just create a cell and run t...
well well well
nocturne mural
well well well
!rm -r /kaggle/working/program_ml/program_ml/
oak edge
```!rsync -a /kaggle/working/program_ml/program_ml/ /kaggle/working/program_ml/
...
i think it worked maybe
oak edge
```!rsync -a /kaggle/working/program_ml/program_ml/ /kaggle/working/program_ml/
...
-kaggle
nocturne mural
i think it worked maybe
Now run the cell to start Applio.
oak edge
let's see
oak edge
Now run the cell to start Applio.
silent goblet
how can i fix lag on ai voice changer like voice is lagging so much, when i used gtx 1050 i didnt have any problems and now with rx 580 its not working
i have ryzen 5 3600 as cpu also
sick mauve
While downloading i got ''too many requests'' from github, not letting me download 3 (pretrain) files, do i just wait or...
viral mason
The only pretrain you should be downloading is legacy core, the others cause models to sound metallic or have problems with them in general
sick mauve
The only pretrain you should be downloading is legacy core, the others cause mod...
its the autodownload from MMVCServerSIO (deiteris okada)
[WeightDownloader] Failed to download or verify pretrain/crepe_tiny.pth
[WeightDownloader] 429, message='Too Many Requests'
viral mason
Ohh
I thought you were downloading pretrains for training voice models my bad
That's odd tho it didn't work
low shard
how can i fix lag on ai voice changer like voice is lagging so much, when i used...
why would you ever go from a gtx 1050 to an rx 580? A Modern Nvidia GPUs is better for AI
and also, there are many different RVC Realtime Voice Changer programs
viral mason
I would be heart broken if someone downgraded like that
low shard
I would be heart broken if someone downgraded like that
I mean, the RX 580 is technically better than the gtx 1050, but AI support is worse on AMD generally speaking, they are both old GPUs, it would just be better to get a modern rtx serie
icy vessel
how to resume training in the applio colab, the applio colab guide shows an outdated or different version from the current one
sick mauve
check https://docs.aihub.gg/realtime-voice-changer/local/deiteris-w-okada-fork/#...
it was a github rate limit
just had to wait
viral mason
I mean, the RX 580 is technically better than the gtx 1050, but AI support is wo...
That's what I meant by downgrading, just for the ai support since you'd have only 2 options instead of all 3
low shard
it was a github rate limit
i mean yeah, you could either wait or use the already zipped version
weird that you got limited tho, did you try to delete and reinstall the program multiple times?
sterile yacht
you need DML version
Where can I download this?
simple ore
Where can I download this?
-rt
simple ore
3rd link
sick mauve
is there text to voice?
lost ember
is there text to voice?
cryin for ts
low shard
is there text to voice?
TTS?
low shard
how to resume training in the applio colab, the applio colab guide shows an outd...
some parts of the applio ui colab guide on the docs have just been updated, maybe try with the sync with google drive cell
low shard
yeah, ig
in realtime for calls ? since I saw you talking about rvc realtime voice changers before
sick mauve
in realtime for calls ? since I saw you talking about rvc realtime voice changer...
sure
low shard
sure
Wokada nor any RVC Realtime Voice Changers can do TTS in realtime
RVC is STS, not TTS
for doing realtime tts, you'd need to check https://docs.aihub.gg/tts/realtime-tts/
sick mauve
ok ty
low shard
for doing realtime tts, you'd need to check https://docs.aihub.gg/tts/realtime-t...
oops i did a typo in the link, it's now fixed
untold anchor
yo, so i got everything working, it's just that my friend hears my normal voice before the egirl voice. any fix???
pure quiver
Hi is it possible to use voice changer on a Mac?
keen wraith
so im trying to upload the voice changer and it wont allow me saying the file doesnt contain beatrice
knotty moth
how can i fix lag on ai voice changer like voice is lagging so much, when i used...
why not just get an RTX 2060 or higher depending on ur budget?
please read the spec requirements in the guide
silent goblet
why would you ever go from a gtx 1050 to an rx 580? A Modern Nvidia GPUs is bett...
because i got it for games im getting rtx 3070ti next year
i was curious just because gtx 1050 did a good job and rx not working at all
low shard
i was curious just because gtx 1050 did a good job and rx not working at all
if you are using wokada tg-develop or deiteris fork dml windows version, it should work
but be aware that nvidia gpus have better ai support
swift thunder
Does anyone know how I can see the sample rate of a model I've already created? It's a bit old.
supple garnet
I can’t hear myself or da model
supple garnet
I can’t hear myself or da model
Any help I can get from anyone
viral mason
What is their differences if i may ask
still happening with the single repeating link
didn't mean to reply to u sorry
viral mason
delete the notebook and retry from scratch
doing what u told me didn't help
knotty moth
still happening with the single repeating link
that happened to me, after trying to delete and recreate my ngrok account, it shows all same public urls
rn I have moved to using localtunnel, though the upload speed in the filebrowser may be slow as hell
knotty river
anyone know good real-time voice changer for linux?
hallow thistle
anyone know good real-time voice changer for linux?
!howtoask
hallow thistle
What is your PC GPU? W-Okada realtime voice changer works best with a dedicated GPU like NVIDIA GeForce RTX.
brittle wing
guys why cant my voice work on RVC models but it works on beatrice ones
i have an nvidia rtx 4050
and an amd ryzen 7
hallow thistle
i have an nvidia rtx 4050
No fixes for the outdated "original" W-Okada version. For better W-Okada, try the Deiteris or Tg Develop W-Okada fork instead. https://docs.aihub.gg/realtime-voice-changer/local/deiteris-w-okada-fork/#download-nvidia-on-windows
brittle wing
would you recommend the deiteris or the tg develop one?
hallow thistle
would you recommend the deiteris or the tg develop one?
Deiteris for the most stable (b2332). If you want the very more recent features that not featured in Deiteris, I'd say Tg Develop.
maiden condor
why does ai voice sounds like from a cave?
maiden condor
i reinstalled vb drivers and now all good
stray plume
Voice Changer no working
meager coyote
res is to high even tho i dont have a outdated version and the ping is to high
viral mason
res is to high even tho i dont have a outdated version and the ping is to high
which one are u using? wokada deiteris, tg fork, or vonovox?
meager coyote
wokada
viral mason
there are multiple wokada
meager coyote
uhh
viral mason
there's the original, deiteris, and tg fork
meager coyote
No fixes for the outdated "original" W-Okada version. For better W-Okada, try th...
i downloaded from here
dml b2332
viral mason
u downloaded this one right?
meager coyote
yeah
viral mason
should work just fine for u then, what settings do u have?
viral mason
oh
meager coyote
but i changed cpu and gpu
spice glen
@low shard hey there idk if u still remember me but how u doing?
i want to know where i can find a model of this voice
meager coyote
i cant change it to rmpve but i have rmpv_onnx
is that fine
still getting 16k ms
viral mason
i cant change it to rmpve but i have rmpv_onnx
that's due to being on amd
I am on nvidia
viral mason
Voice Changer no working, what to do
did u download it from a yt tutorial?
stray plume
did u download it from a yt tutorial?
yes
viral mason
it's really old and outdated then, even if the video was from like a month ago
stray plume
did u download it from a yt tutorial?
How do I send a screenshot?
viral mason
How do I send a screenshot?
I'd give u the ability but I don't have mod powers :p
so u can't
but what gpu do u have?
stray plume
I'd give u the ability but I don't have mod powers :p
Can I write you a private message?