It keeps saying error
#✨│ai-help
brittle wing
help here how many epoch should i use
maiden mesa
What’s GPU
violet heron
What’s GPU
The graphics card in your computer
violet heron
I’m on my phone
And what are you using to make ai covers
violet heron
The <@1138318590760718416> bot
Try using Ilaria RVC Zero it work on mobile try that instead
-spaces
maiden mesa
Try using Ilaria RVC Zero it work on mobile try that instead
Is it an app?
@violet heron
rare gobletBOT
Ayo? @maiden mesa level 2 !!! 
violet heron
Is it an app?
Website
maiden mesa
I’m confused on the rvc zero
maiden mesa
Ok
I was trying to use CASEOH ai
It’s not on the web
I didn’t think making ai covers was hard
violet heron
It’s not on the web
Do you see the “Download Model” tab
That is where you put the download link for the model
So you can use it
https://docs.google.com/document/d/1YbXcLFPaGjhOdG5NFkK3QrucCEpHZBwFUxkeMO8aB18/edit if you really need a guide somehow
tawdry ginkgo
guys how do I solve this for the sake of god
rare gobletBOT
Ayo? @tawdry ginkgo level 2 !!!
tawdry ginkgo
Im downloading a new NVIDIA driver but idk
because I have a NVIDIA GPU so im really confused
so maybe the problem is some driver?
pastel oak
because I have a NVIDIA GPU so im really confused
i mean it says no nvidia driver found so you havent installed your gpus drivers yet
go to nvidia page look up your gpu and install xd
violet heron
i mean it says no nvidia driver found so you havent installed your gpus drivers ...
The person seems to be using colab
violet heron
guys how do I solve this for the sake of god
Are you using a colab?
tawdry ginkgo
Are you using a colab?
yep
I already downloaded the driver, so lets see if this will gonna solve the problem
no still no working...
idk what to do
violet heron
I already downloaded the driver, so lets see if this will gonna solve the proble...
It won’t
Colab uses googles GPU
Make sure your connected to a runtime (top right)
If your trying to make a cover I highly recommend Ilaria RVC zero instead
tawdry ginkgo
If your trying to make a cover I highly recommend Ilaria RVC zero instead
srry but can you send a link of that RVC?
nvm already found it
tawdry ginkgo
If your trying to make a cover I highly recommend Ilaria RVC zero instead
but that means now that I need to download both vocals and instrumentals, dang, like in aicovergen that was more easy...
now this is appearing in aicovergen: "Can't connect to GPU backend.
Currently, it is not possible to connect to a GPU due to Colab usage limits. Learn more
For more access to GPUs, purchase Colab compute units with Pay As You Go."
like are you fucking telling me that I need to PAY to still use this google colab? bruh capitalism
violet heron
now this is appearing in aicovergen: "Can't connect to GPU backend.
Currently, i...
Use an alt
tawdry ginkgo
an alt acc right?
wraith bobcat
Umm i am new so whats rvc?
violet heron
an alt acc right?
yes
wraith bobcat
and how can i download this voice changer?
violet heron
Umm i am new so whats rvc?
RVC is a free and open source voice-to-voice ai program interfaced with gradio
knotty moth
but that means now that I need to download both vocals and instrumentals, dang, ...
for RVC inference, use Ilaria RVC zero but it needs an input of isolated deverbed vocals
if not yet, do isolate & dereverb in UVR UI https://huggingface.co/spaces/TheStinger/UVR5_UI
wraith bobcat
RVC is a free and open source voice-to-voice ai program interfaced with gradio
Thanks bt how can o download this?
violet heron
Thanks bt how can o download this?
Are you trying to use the voices for covers or real time?
wraith bobcat
i want for Youtube recording villager voice
tawdry ginkgo
for RVC inference, use Ilaria RVC zero but it needs an input of isolated deverbe...
alr ty
rare gobletBOT
Ayo? @tawdry ginkgo level 3 !!!
violet heron
i want for Youtube recording villager voice
Do you need to convert the voice in real time or is just audio files converting fine
wraith bobcat
real time
knotty moth
real time
then you need the voice changer app, I'd suggest this https://github.com/deiteris/voice-changer
wraith bobcat
Thanks
tawdry ginkgo
Use an alt
but thats kinda sucks because if this acc reach the colab usage limit, I'll have to switch again to another acc and like, I will have to create 1899870 google accounts 😭
to use this google colab again, so yeah...
wraith bobcat
k
knotty moth
but thats kinda sucks because if this acc reach the colab usage limit, I'll have...
I'd rather use colab spaces for training, yet inference is doable in ilaria RVC zero or local fork even with just decent modern laptop CPU
violet heron
like yeah RVC is free, but theres a limit you know, im tired of this capitalism
Only on colab*
Once again I already said to use ilaira RVC zero
tawdry ginkgo
okay tysm anyways guys, Im going to use this illaria rvc for now
brittle wing
Could I do an AI cover from my phone or is it too much?
rare gobletBOT
Ayo? @brittle wing level 1 !!!
proper shale
Could I do an AI cover from my phone or is it too much?
yeah, just do it on websites
https://docs.google.com/document/d/1YbXcLFPaGjhOdG5NFkK3QrucCEpHZBwFUxkeMO8aB18/edit guide for Ilaria RVC Zero, basically easiest and fastest way to make covers
proper shale
not sure why that happens
proven hill
whats wrong with abortions
tawdry ginkgo
whats wrong with abortions
idk I just cant make the cover I want 😦
because of this
oh nvm now I can? lol
obtuse holly
-colab
stone shale
how can i add voices to b2139DirectML version?
i didn't see any voices like the client one
wispy lodge
how can i add voices to b2139DirectML version?
You should upload them with menu in edit button here
stone shale
You should upload them with menu in edit button here
thanks, and now how can i edit chunk like the client one? here is like 450.0ms and extra 0.5s different than client one
wet zealot
my voice comes out in a stutter, where do I use the -rt command, they told me to use this command exactly, but I don't know where to use it. (voice.ai)
wispy lodge
thanks, and now how can i edit chunk like the client one? here is like 450.0ms ...
It may be easier to refer to the chart for RVC realtime which uses the same units. Sample length (in seconds) corresponds to chunk (in milliseconds), and extra corresponds to extra
https://rentry.co/RVCRealtimeGuide#gpu-chart-for-sample-length-extra-delay-quality
wispy lodge
my voice comes out in a stutter, where do I use the -rt command, they told me to...
There's no support for voice.ai
wet zealot
There's no support for voice.ai
Where can I get voice.ai?
wispy lodge
There are open source alternatives like w-okada voice changer and rvc realtime for realtime voice conversion
stone shale
It may be easier to refer to the chart for RVC realtime which uses the same unit...
thank you so much, i have one question.. this version MMVCServerSIO_win_onnxdirectML-cuda_v.1.5.3.18a.zip
why this version who have the client don't show my gpu?(7900XT) just the other one who open the browser
rare gobletBOT
Ayo? @stone shale level 1 !!!
wet zealot
There are open source alternatives like w-okada voice changer and rvc realtime f...
Is there a rvc real time video showing how to download this program?
violet heron
Is there a rvc real time video showing how to download this program?
YouTube videos are often outdated and spread misinfo
-realtime
azure marshBOT
-realtime
This interaction has expired, use the command
/guides realtimeif you wish to see it again.
sharp solstice
-realtime
azure marshBOT
-realtime
This interaction has expired, use the command
/guides realtimeif you wish to see it again.
wispy lodge
thank you so much, i have one question.. this version MMVCServerSIO_win_onnxdire...
It's how this version works. It doesn't list AMD/Intel GPUs so you have to choose a corresponding GPU number
sharp solstice
-colab
sharp solstice
-rvc
AI HUB Docs
azure marshBOT
🎙️ Realtime Voice Changer (Online)
indigo pilot
Does rvc have amd support yet?
signal willow
is there a program to get nice sounding text to speech for rvc models? one that is free and doesn't use a default tts model and then converts that audio into your character/model using rvc? i want one that just uses the rvc model only, no in-between with a default tts model
violet heron
Does rvc have amd support yet?
No so it would use CPU instead which will be slower, if your talking about w-Okada real time yes you would
patent trellis
-colab
low shard
is there a program to get nice sounding text to speech for rvc models? one that ...
It's not possible, rvc is made for speech to speech (STS), not tts natively, so unless having in between a tts model, its impossible to do that
pliant agate
how do u use the pretrain models in #1235952130855010365
low shard
It's not possible, rvc is made for speech to speech (STS), not tts natively, so ...
if you are looking especially for good text to speech (tts), maybe you could try gpt so vits but its limited to only english, chinese & japanese, and u can't use rvc models here
low shard
how do u use the pretrain models in <#1235952130855010365>
the pretrains are usable only while training, check https://docs.google.com/document/d/1j9J8A8Oop9bMOHmCs3jDXzPujuD6TQ0Q396rJ0MyuIc/edit
signal willow
It's not possible, rvc is made for speech to speech (STS), not tts natively, so ...
well then what is the best way to make it sound like the model
low shard
well then what is the best way to make it sound like the model
either use the built in edge tts in the forks such as applio and ilaria rvc (ilaria rvc zero & mainline versions both have it) and play around with the settings, or generate a tts audio file with 11labs and use that as an input in rvc
rare gobletBOT
Ayo? @signal willow level 1 !!!
signal willow
either use the built in edge tts in the forks such as applio and ilaria rvc (ila...
i use applio, what do you mean by edge tts?
low shard
i use applio, what do you mean by edge tts?
the Applio TTS, uses Microsoft's Edge TTS via api, you can just use it and play around with the settings till you get the best results
signal willow
the Applio TTS, uses Microsoft's Edge TTS via api, you can just use it and play ...
what is the best tts voice to use most of the time?
low shard
what is the best tts voice to use most of the time?
there isnt a best one, the edge tts models are different, i suggest you putting one that is the same gender and language as the rvc one
signal willow
there isnt a best one, the edge tts models are different, i suggest you putting ...
how do i give it emotion? it just always ends up sounding bland when they speak
low shard
how do i give it emotion? it just always ends up sounding bland when they speak
Edge TTS is good for being multilingual, but isn't much good with emotions maybe you could try to play with the speed, but if you care about emotions much its better you generate the tts audio file with 11labs and use it as an input in applio (11labs has 10k free characters per month)
signal willow
can i upload models to eleven labs?
low shard
can i upload models to eleven labs?
like use the rvc models in 11labs? i dont think so
low shard
Edge TTS is good for being multilingual, but isn't much good with emotions <:nai...
what i meant was, make a tts audio in 11labs with a voice with same language and gender for free, and then use it in applio where u can use the rvc model
thin stump
can i upload models to eleven labs?
The TTS of elevenlabs and dataset cleaning workflow is closer than you think so you might as well get into the habit of making voice models first
patent trellis
-colab
patent trellis
what do i do to Index File and Voice Model?
rare gobletBOT
Ayo? @patent trellis level 2 !!!
fickle tundra
hey i was trying to open tensorboard from the ai hub docs thing and got this error can someone help me fix it?
the full log with the error if more context is needed
violet heron
what do i do to Index File and Voice Model?
Do you want to use the voices in real time or for covers?
fickle tundra
hey i was trying to open tensorboard from the ai hub docs thing and got this err...
pls help
fickle tundra
nvm fixed it
brittle wing
Hey! Good night, I am quite new into this hugging face models and things to do ai covers
And I want to make my first one
But I don't know quite well how can I do it
Can someone help me?
fickle tundra
hey so i pressed train model and its not saying stop training, if i click on it its still gonna stop right?
command line proof its running
thin stump
Hey! Good night, I am quite new into this hugging face models and things to do a...
you can find the models in weights.gg like the guide says
thin stump
hey so i pressed train model and its not saying stop training, if i click on it ...
and yes
brittle wing
you can find the models in weights.gg like the guide says
Thanks!
thin stump
Thanks!
damn I don't like the new look of the site. It used to be so simple so bare with it
brittle wing
damn I don't like the new look of the site. It used to be so simple so bare with...
Ok! I will 👍
misty elk
-colab
patent trellis
-colab
wraith bobcat
Bro
I have downloaded the client
But I want to listen my own voice in speaker how can I do that
It looks like this
@dusty rampart
rare gobletBOT
Ayo? @wraith bobcat level 2 !!!
gray tinsel
idk what i did but it's working beautifully now 😂
gray tinsel
But I want to listen my own voice in speaker how can I do that
i use discord voice/audio test in user settings to hear myself
wraith bobcat
i use discord voice/audio test in user settings to hear myself
how can i do that
gray tinsel
how can i do that
i have it setup like this, headset out, virt cable in
dawn tendon
Yes
wraith bobcat
;-;
dawn tendon
I guess
wraith bobcat
fix u know?
dawn tendon
Idk how
wraith bobcat
;-;
dawn tendon
Can't you help?
wraith bobcat
idk i m new
gray tinsel
i headset input, and vb cable output
fickle tundra
and yes
hey so i clicked on it and it didnt stop, which is a problem because i was following the ai hub doc and set it to 1000 epochs
steel forge
<@159310300275802112> pls help me
Please don't ping randomly
past trench
How can I put an old song sung by an old artist on an artist who has sung the same song but newly
That is, artificial intelligence changes the singer
Pls help 🖤
wraith bobcat
kk
steel forge
But I can help, does your w-okada have changing numbers in the box floating above the avatar?
wraith bobcat
PLS HELP
wraith bobcat
NAH BRI IDK
steel forge
You have nvidia gpu right
steel forge
OK so
wraith bobcat
4gb ram
steel forge
W-okada requires a GPU kinda
It'll run in integrated graphics but its gonna like, die
wraith bobcat
;-; so i cant run ?
steel forge
It won't sound like a convincing voice lol
wraith bobcat
i want to make video for minecraft villager
steel forge
So you just want to add audio on top of a video?
wraith bobcat
yes
steel forge
I use Applio at the top, but they all should work
steel forge
You load the villager model into it, then give it the input vocals in an audio file. Then you can convert it
steel forge
i have to dwonload all these?
No, it's all web based
wraith bobcat
umm so? on web?
steel forge
Go to applio at that top of the list
wraith bobcat
i want to transfer my voice to villager voice but i hava added the voice the voice not coming out
see screenshare in vc
past trench
How can I put an old song sung by an old artist on an artist who has sung the same song but newly
That is, artificial intelligence changes the singer
Pls help 🖤
wide hollow
Hello, I have a problem when I try to generate a cover with AIcoverGen-WebUI. I only get the artist's voice but not the instrumental. If anyone can help me, I would appreciate it.
gray tinsel
this made me laugh, it immediately bugged out when i said "whaaaaaaaaaaat"
can i ignore the third thing?
wispy lodge
Third thing is a formant shift
brittle pelican
-colab
naive fossil
which paperspace rvc works now?
proven hill
which paperspace rvc works now?
only ilaria rvc
wise saddle
glad veldt
just found it doesnt work with models that dont have .index
brittle wing
any help ? ModuleNotFoundError: No module named 'fairseq'
rare gobletBOT
Ayo? @brittle wing level 5 !!!
brittle wing
pip install fairseq doesnt work either
crude bolt
sorry i dont use gradio - since many here told me there are better alternatives. I have no experience with Gradio.
rare gobletBOT
Ayo? @brittle wing level 1 !!!
crude bolt
https://applio.org or native rvc - but since its harder to setup i would recommend first one.
(they both use same techniques, applio is based on native rvc code).
crude bolt
in general on a collab page - errors magically appear and dissapear in time.
brittle wing
Anyone help? I'm trying to run RVC Mainline in Kaggle using instructions given from Cauthess. Ngrok token and custom pretrains applied. All I get is "We can’t connect to the server" error ||(this also happens on Chrome)|| on all 3 links given despite an Ngrok token is applied. Thanks.
crude bolt
i already ran feature extraction
you could try to do the steps with another model you have used succesfully in the past. If it also errors you know its not on your side. If it passes, you might check the model again that you downloaded.
or well since feature extraction is about doing it with wave data , you can check the audio file that u used.
sorry for mixing the term model and audio file.
boreal sage
you could try to do the steps with another model you have used succesfully in th...
switched to applio and it works
odd shale
been using easy gui for almost a year lol
Have you got any problem using Easy GUI till now?
unborn owl
Have you got any problem using Easy GUI till now?
no problems i had, i used the new one and it wasnt expired like the old one
i like the new one
thick dagger
-colab
thick dagger
which google colab i can use for free ?
rotund prawn
-hf
obtuse stream
how can i solve this
C:\Applio-3.2.1\env\lib\site-packages\torch\autograd_init_.py:251: UserWarning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance.
violet heron
https://applio.org or native rvc - but since its harder to setup i would recom...
Unzipping and opening a bat file?
Instead of unzipping and opening a different bat file?
knotty moth
local or cloud? if latter, I'd recommend this https://docs.google.com/document/d/1YbXcLFPaGjhOdG5NFkK3QrucCEpHZBwFUxkeMO8aB18/edit
grim sphinx
how to download this apk?
gray tinsel
is rtx 3080 laptop gpu good for 30ms latency, or should i find a 4090 desktop? my 1070 gets 1-2s delay
rvc
night moss
whats good settings for applio?
the voices always sound kinda "shaky"
rare gobletBOT
Ayo? @night moss level 2 !!!
wispy lodge
is rtx 3080 laptop gpu good for 30ms latency, or should i find a 4090 desktop? m...
You can reduce that delay to like 840-860ms...
Change sample length to 0.18-0.2
gray tinsel
You can reduce that delay to like 840-860ms...
of interest thanks, but it gets choppy
oop i see, takes a bit to warm up
odd error, but models are working fine
dense moon
Im using applio cloud rvc, im trying import an audio but it only comes out as nothing
Is there something i could do to fix it?
grim sphinx
There is no apk
n how to use this web?
grim sphinx
Are you trying or make covers or use in real time
use in real timee
violet heron
use in real timee
Can’t on mobile
Unless you count colab but you wouldn’t be able to on any apps
grim sphinx
Can’t on mobile
i have pc
rare gobletBOT
Ayo? @grim sphinx level 3 !!!
grim sphinx
What’s your GPU
rx 560
violet heron
rx 560
https://github.com/deiteris/voice-changer/releases/tag/b2139 if your on windows download the dml version
grim sphinx
https://github.com/deiteris/voice-changer/releases/tag/b2139 if your on windows ...
thank youu
grim sphinx
https://github.com/deiteris/voice-changer/releases/tag/b2139 if your on windows ...
this is okada?
grim sphinx
is there a slight delay when changing the voice?
dense moon
Is there anything in advanced settings of applio rvc i should change for a female character to sing bury the light?
As for pitch i chose either 16 or 18
violet heron
is there a slight delay when changing the voice?
Yes, there will be a delay no matter what
violet heron
Yes, there will be a delay no matter what
Depending on your settings your delay will vary
grim sphinx
A fork of it
Is there any way to reduce the delay?
violet heron
Is there any way to reduce the delay?
Increasing your chunk and decreasing extra
It will affect quality tho
The fork I sent should be better on AMD then normal w-Okada so it should work better
dense moon
uhh whats the fo curve??
is this something new thats needed? Says i should a drop a file but what?
grim sphinx
Increasing your chunk and decreasing extra
aaa thank u for helping me
plush saffron
Hey, is it possible to use RCV in real time in Discord? Thank you!
gray tinsel
Hey, is it possible to use RCV in real time in Discord? Thank you!
yes, similar setup should work
violet heron
is this something new thats needed? Says i should a drop a file but what?
Don’t need to
That’s for uploading a custom one
dense moon
Don’t need to
Alright but.. i tried converting but it came out as just an empty audio..
violet heron
Alright but.. i tried converting but it came out as just an empty audio..
Did you select a model?
dense moon
Im using applio rvc btw
violet heron
Im using applio rvc btw
applio moment
dense moon
Did you select a model?
i just copy pasted i kinda forgotten how to do the actual importing thing.. 💀
But i do have the model
Pth and index
violet heron
i just copy pasted i kinda forgotten how to do the actual importing thing.. 💀
Click refresh if it’s there then select the model
Also make sure your volume is up
rare gobletBOT
Ayo? @dense moon level 4 !!!
violet heron
Send a screenshot
dense moon
Send a screenshot
violet heron
Go to download to import the model
white bough
Anyone else using applio on colab, and the crepe extraction does not work because it doesn't seem to have activated GPU? RMVPE works and then the training uses GPU. Just not the feature extracting part...
cloud pebble
Anyone else using applio on colab, and the crepe extraction does not work becaus...
yeah crepe seems broken, which is unfortunate 😆 every time i try to use applio i run into broken things, keeps pushing me back to other forks
dense moon
omg..
Has anyone who ever successfully made a female character sing bury the light without any artifacts?
spice charm
One message removed from a suspended account.
One message removed from a suspended account.
One message removed from a suspended account.
One message removed from a suspended account.
One message removed from a suspended account.
violet heron
if i switch to linux it will be better?
What’s your GPU
supple ibex
why is this happening when I try to process my dataset?
wispy lodge
Which APU?
wispy lodge
Then use modded version
https://github.com/deiteris/voice-changer
wispy lodge
Not sure I remember, but it's the same vc. There were more optimizations for AMD so it works better on APUs too now
edgy oak
sweet
rare gobletBOT
Ayo? @edgy oak level 9 !!!
violet heron
and i might switch to linux
That will help since your on AMD
edgy oak
That will help since your on AMD
sweet
edgy oak
Depends on the settings
what specs is these?
wispy lodge
That's ryzen 7 5800h igpu
.
edgy oak
That's ryzen 7 5800h igpu
its stronger than my apu so fuck
violet heron
i never used linux
Are you going to install a whole new os just to use a voice changer?
wispy lodge
its stronger than my apu so fuck
igpu is the same
edgy oak
Are you going to install a whole new os just to use a voice changer?
nope there is other reasons
edgy oak
igpu is the same
still strong or?
violet heron
nope there is other reasons
Not really worth it then
edgy oak
cuz i never heared of that
violet heron
is linux bad?
Well your only using it for the voice chnager
edgy oak
I use arch btw :3
ye im so confused mint or arch
violet heron
ye im so confused mint or arch
Mint is good for beginners
edgy oak
a lot uses arch
violet heron
a lot uses arch
:3
edgy oak
and community good
violet heron
and community good
Programerrrr sockssssss
edgy oak
who is better for gaming?
wispy lodge
still strong or?
I mean, you can try. Someone even managed to run it on ryzen 3 3200g which is much weaker
With not the greatest settings, but it works
edgy oak
I mean, you can try. Someone even managed to run it on ryzen 3 3200g which is mu...
25% diff
violet heron
who is better for gaming?
Linux?
edgy oak
Haven’t really used mint so I dunno
ur using arch?
violet heron
ur using arch?
Yeah dual-booted with windows
edgy oak
Yeah dual-booted with windows
i can use both?
edgy oak
And?
i mean i tried it 4 months ago
i remember the updates were every 15+ days
now its good
violet heron
i can use both?
If you dual boot or if you use a different thing to boot of off
edgy oak
so is it worth try again?
edgy oak
If you dual boot or if you use a different thing to boot of off
do i need usb?
cuz i dont have one
iso works?
violet heron
Yeah
wispy lodge
so is it worth try again?
I've already told everything and there are demo videos in readme. Make your own decision
edgy oak
ok what about the performance is it gonna be reduced?
wispy lodge
If you think making rocm work with 5600g is easy, then good luck. Google first
violet heron
ok what about the performance is it gonna be reduced?
Performance on Linux might be higher or lower
For gaming
edgy oak
I've already told everything and there are demo videos in readme. Make your own ...
i mean idk if i switch to linux it will be better?
violet heron
For gaming
For the voice changer it might be hard getting rocm to work
But i I would highly recommend using the w-Okada fork and seeing if that’s good for you
violet heron
AR CH OR MINT??
Mint is easier to learn
violet heron
the best
I’m biased towards arch so I would say that
Try to see online which one would work best for you
Once again, I recommend trying the w-okada fork and seeing if that works better
edgy oak
i had enough using windows
rotund moss
I'm sorry to ask, but is there any way to continue training on the next day with easygui? (I have saved pth successfully because I adjusted the save rate to 20) The training was interrupted due to Google colab's daily free usage limit.
white bough
After going into the code, it seems to me that (Mangio)crepe is unusable with the new version of Applio (3.2.1) because the feature extraction has been moved from GPU to CPU. And it looks like this was done on purpose, but I do not understand why. Should I change an option for it to work?
analog obsidian
After going into the code, it seems to me that (Mangio)crepe is unusable with th...
if you want to use mangio-crepe your only best option is codename's fork, but if you want to continue using applio, you have to try another version where crepe is working (i don't know which version is the last one to support crepe)
im not sure if applio devs are aware of mangio-crepe not working there tho, but you can try reporting this issue in applio's github page
slate bramble
do you guys think i should train my model more or is this already decent? (i was ranting ab roblox)
modern ferry
steep hinge
Hey all, just wondering, what do you all do to improve RVC's tts generation audio? I'ts managed to tune settings to the point where audio comes out sounding pretty decent, but there's constant electronic-like interference during some spoken words, and I can't seem to get rid of it no matter what. Not sure how it's called specifically, i've heard it called flanging or something like that, in case that makes it any clearer.
radiant bison
what software do i use for this
low shard
Hey all, just wondering, what do you all do to improve RVC's tts generation audi...
RVC Is made for Speech To Speech (STS), not tts so of course is not as good at it as gpt so vits, btw what fork are you using
steep hinge
Applio RVC
low shard
what software do i use for this
For what? There's 2 different softwares for ai covers, rvc which u can check in https://docs.aihub.wtf/, and then for realtime ai voice changer for calls that is wokada and u can check in https://rentry.co/voicechangerguide
radiant bison
For what? There's 2 different softwares for ai covers, rvc which u can check in ...
i neeed rvc
low shard
Applio RVC
Applio TTS just uses Microsoft' Edge TTS, which is good for multilingual but not emotions, applio TTS basically works in using an edge tts model for generating a tts output audio and then using an RVC model on that
radiant bison
wait do i need a good pc for it?
low shard
You could play around for the settings till it sounds the best, ofc rvc isn't very good for tts
low shard
You could play around for the settings till it sounds the best, ofc rvc isn't ve...
Also, you can try generating an audio tts file by eleven labs for free (10k characters per month freely) and using it as an input in rvc, maybe that would work better
low shard
wait do i need a good pc for it?
For doing it locally yes, else use Ilaria RVC Zero that is the fastest free way for RVC on cloud, or the Google colab like applio in the https://docs.aihub.wtf/
radiant bison
thanks
rare gobletBOT
Ayo? @radiant bison level 1 !!!
radiant bison
i cant find the site for the cloud
thin stump
Hey all, just wondering, what do you all do to improve RVC's tts generation audi...
It's the ringing noises. That's part of the model itself #🧬│ai-chat message so that's why you don't pick up strays from weights.gg and only choose from #1175430844685484042
radiant bison
in the google doc
knotty moth
Hey all, just wondering, what do you all do to improve RVC's tts generation audi...
seems model quality issue, you can try adjusting index rate and consonant protection if possible
radiant bison
oh so its just online no download
steep hinge
Applio TTS just uses Microsoft' Edge TTS, which is good for multilingual but not...
emotions in voice are not the issue, its fine by me, the problem is the audio quality itself. Like, 80% of stuff generates properly, but the rest 20% have some sort of electric-like interference, like as if they were spoken into cheap 5$ microphone
Like in the first "investment" word
theres some weird artifacting or something
low shard
oh so its just online no download
Yes, as you don't have a good PC (from what you said) so you can do it online
steep hinge
I can't quite describe it properly
low shard
emotions in voice are not the issue, its fine by me, the problem is the audio qu...
Could also be a problem of the RVC voice model quality (like the model could be low quality), I can't really listen to the audio as it's almost 3 am but maybe you could see if you still get it with other models
patent trellis
-colab
vital acorn
bruh colab sucks
can someone please give me any site but colab thats free and allows custom ai models
i have one ready to use but i give up on colab
patent trellis
-colab
flat patrol
How do I make a model?
violet heron
How do I make a model?
-docs
rare gobletBOT
Ayo? @flat patrol level 1 !!!
amber solar
I keep on getting a error when trying to make a cover
gray tinsel
where should i put response threshold? previous messages are conflicting
violet heron
I keep on getting a error when trying to make a cover
!howtoask
patent trellis
brittle wing
please help.. getting this error on mac when i want to run applio: ModuleNotFoundError: No module named 'regex'
pip install regex didn't help
wispy lodge
You don't need chunk at 2.7s. Your sound will be delayed by 2.7 seconds because of that
Set it to 512ms
grim sphinx
oke i'll try it
grim sphinx
You don't need chunk at 2.7s. Your sound will be delayed by 2.7 seconds because ...
when I speak the voice will be delayed
wispy lodge
when I speak the voice will be delayed
Well, yes, because chunk controls the delay
But your GPU should also be fast enough to process smaller chunks
wispy lodge
mac settings m1 chip
If res goes above buf, your gpu does not keep up
Reduce extra
afaik, known working settings for original version are rmvpe_onnx, chunk 256 and extra 8192
river thicket
could u recommend a good modek>
rare gobletBOT
Ayo? @river thicket level 2 !!!
river thicket
model?
wispy lodge
What kind of?
river thicket
like a girl type
wispy lodge
This one is good #1229978221655756911 message
obtuse stream
anyone help me how to solve this?
C:\Applio-3.2.1\env\lib\site-packages\torch\autograd_init_.py:251: UserWarning: Grad strides do not match bucket view strides. This may indicate grad was not created according to the gradient layout contract, or that the param's strides changed since DDP was constructed. This is not an error, but may impair performance.
grad.sizes() = [64, 1, 4], strides() = [4, 1, 1]
bucket_view.sizes() = [64, 1, 4], strides() = [4, 4, 1] (Triggered internally at C:\actions-runner_work\pytorch\pytorch\builder\windows\pytorch\torch\csrc\distributed\c10d\reducer.cpp:334.)
Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass
simple onyx
How many epochs and save frequency should I use if my audio file is at 10 minutes and 24 seconds? I'm using RVC V2 Disconnected.
rare gobletBOT
Ayo? @simple onyx level 1 !!!
low shard
How many epochs and save frequency should I use if my audio file is at 10 minute...
There isn't a right amount of epoch, check https://docs.aihub.wtf/rvc/resources/epochs--tensorboard/
And for save frequency I'd say 15 or 25
brave compass
hey guys
rare gobletBOT
Ayo? @brave compass level 1 !!!
brave compass
how to fix voice changer native client wait web infinite:(?
unborn pulsar
is rvc gui outdated?
brave compass
i guess no
brave compass
is rvc gui outdated?
do u know how to fix it:(
unborn pulsar
do u know how to fix it:(
nope
low shard
is rvc gui outdated?
this https://github.com/Tiger14n/RVC-GUI ? yes
check the updated ones in https://docs.aihub.wtf
unborn pulsar
this https://github.com/Tiger14n/RVC-GUI ? yes
which one should i use instead?
rare gobletBOT
Ayo? @unborn pulsar level 1 !!!
unborn pulsar
oh ok
low shard
how to fix voice changer native client wait web infinite:(?
ask for help in #🔍│help-w-okada or #1192011222023950368
low shard
which one should i use instead?
https://docs.aihub.wtf/ either applio or mainline
unborn pulsar
okok thanks
low shard
also be sure u have a good pc for doing it locally
unborn pulsar
also be sure u have a good pc for doing it locally
i read the guide, it said the folder should be in c drive? or it doesnt matter in another drive
inland folio
i read the guide, it said the folder should be in c drive? or it doesnt matter i...
Where guide Bro... I'm newbie
unborn pulsar
https://docs.aihub.wtf/ either applio or mainline
https://docs.aihub.wtf/ click on rvc > local > applio
low shard
i read the guide, it said the folder should be in c drive? or it doesnt matter i...
for applio its suggested that
i dont do it locally btw as my pc isnt good
low shard
Where guide Bro... I'm newbie
guide for what
inland folio
guide for what
Guide for Ai vocal, I've forgotten how to do it
low shard
Guide for Ai vocal, I've forgotten how to do it
realtime voice changer for calls, or ai cover?
inland folio
realtime voice changer for calls, or ai cover?
My bad... just ai cover 😄
rare gobletBOT
Ayo? @inland folio level 1 !!!
inland folio
dw, do you got a good pc?
No... but i have done Kafka ai cover before
low shard
No... but i have done Kafka ai cover before
Yes there's cloud ways,
- Ilaria RVC Zero The fastest free cloud rvc
- Applio Colab that uses the free daily limited (4 hours) of T4 GPU
inland folio
i can't find model loader tag...
low shard
i can't find model loader tag...
wdym
low shard
It not hugging ai ?
what
late fiber
how do i make voice model
finite lintel
@timid olive ses eğitemiyorum dm bakar mısın
timid olive
finite lintel
-colab
rare gobletBOT
Ayo? @finite lintel level 1 !!!
finite lintel
hangisi en iyisi @timid olive
misty elk
-colab
vague sandal
how do i make the voice change client play a downloaded song ?
violet heron
how do i make the voice change client play a downloaded song ?
Use RVC instead of w-Okada for covers
pearl pollen
afaik, known working settings for original version are rmvpe_onnx, chunk 256 and...
I am haveing the same exact issues I just chaged those settings and mine is still just tons of static and delay
these are my settings
pearl pollen
mac settings m1 chip
for some reason its not letting me upload a photo, but my settings are the exact same as this persons just with the chunk at 256 and the extra at 8192
rare gobletBOT
Ayo? @pearl pollen level 1 !!!
low shard
how do i make voice model
check https://docs.aihub.wtf/essentials/how-to-make-voice-models/, if you got a good pc use local else cloud
wispy lodge
for some reason its not letting me upload a photo, but my settings are the exact...
You can upload as screenshot now
brittle wing
how can i seperate main and backing vocals as seperate tracks?
proper shale
how can i seperate main and backing vocals as seperate tracks?
use a karaoke separation method, like for example UVR's BVE (can be found on MVSep as well)
or maybe just use this
https://huggingface.co/spaces/TheStinger/UVR5_UI
go to VR Arch, use UVR-BVE, upload audio, separate, be happy
brittle wing
use a karaoke separation method, like for example UVR's BVE (can be found on MVS...
i dont understand the settings? like the agression and window size and stuff
proper shale
i dont understand the settings? like the agression and window size and stuff
these r fine, keep as is
id say mess with aggressiveness if the result wasnt that good/too aggressive filtering
tawdry ginkgo
guys what is a dataset? cause like Im a beginner so
violet heron
guys what is a dataset? cause like Im a beginner so
In the case for RVC it’s audio used to train a model
twin moat
has anyone had a similar error, as I'm guessing there should be something displayed in the logs afterwards?
(RVC v2 Disconnected on google collab)
tawdry ginkgo
In the case for RVC it’s audio used to train a model
👌
violet heron
In the case for RVC it’s audio used to train a model
Good datasets are in .wav and are very clear without any reverb or background audio
Usually 10-20 minutes
thin stump
👌
read this part. It'll help you with your separation:
https://rentry.co/RVC-dataset-RX11#getting-a-dataset
https://rentry.co/RVC-dataset-RX11#preparing-the-dataset-through-musicsfx-removal
SCRfilms has the spotify rip for you so
thin stump
dull dagger
is there a way to use rvcs speech recognition model as a timestamper when syllables start?
low shard
is there a way to use rvcs speech recognition model as a timestamper when syllab...
So like speaker diarization (separate the speakers timestamp for when there's multiple people talking in a file) or like using it to make a subtitle?
In any cases, I don't think you can but there are other AIS you can use
worn river
Is there another train voice model site in addition to Applio, Weights, Astra and Colab?
low shard
Is there another train voice model site in addition to Applio, Weights, Astra an...
Also Kaggle, which is a free cloud computing service for data science, that gives for free 30 hours weekly of P100 & T4x2
-kaggle
low shard
Also Applio isn't a 'site', it's a fork (modified version) of rvc, there are also other versions such as Ilaria RVC Mainline, or the original Version called Mainline
worn river
Also Kaggle, which is a free cloud computing service for data science, that give...
sorry, but is a site or a software/open source?
rare gobletBOT
Ayo? @worn river level 2 !!!
low shard
sorry, but is a site or a software/open source?
kaggle is a site, for cloud computing power used mostly for data science, basically like google colab but gives more resources and its a bit harder to use
mental wagon
What is the best rvc UI to use to clone voices, because the one i use is severely outdated and only supports harvest and dio as an extraction method
low shard
What is the best rvc UI to use to clone voices, because the one i use is severel...
there isn't technically a best rvc, as i think you are doing it locally, id suggest to try either:
- Mainline (Original version)
- Applio (fork)
- [Ilaria RVC Mainline](#📰│dev-updates message) (fork, there's no guide for it as of right now)
obtuse bane
Hi everyone it's been a while!
First time on apple silicon and after a test run on rvc applio, I've got an error when trying to create the index file: Failed to train index: need at least one array to concatenate
I don't know if it's macos related or not... the last time I used rvc was one year ago and I saw a lot of changes
mental wagon
there isn't technically a best rvc, as i think you are doing it locally, id sugg...
thank you
rare gobletBOT
Ayo? @mental wagon level 1 !!!
mental wagon
there isn't technically a best rvc, as i think you are doing it locally, id sugg...
yeah i should of told you i was trying to do it locally so you didnt need to guess. but nonetheless THANK YOU
low shard
yeah i should of told you i was trying to do it locally so you didnt need to gue...
you're welcome
pearl pollen
You can upload as screenshot now
I'm just getting tons of static and a ton of delay and whenever I try to talk but sometimes when it dose get a voice out it isnt anything you can understand and sounds like complete gibberish
I'm on an MacBook Air
with a M1 chip
obtuse bane
with a M1 chip
oh you're on apple silicon too! If I understand correctly, you have a glitchy audio. I've made some tests yesterday and I found that harvest or dio were excellent when it comes to speed and quality
I'm on the m2 chip tho
pearl pollen
oh you're on apple silicon too! If I understand correctly, you have a glitchy au...
yea, i have a mic though but that dosent seem to change anything lol
pearl pollen
oh you're on apple silicon too! If I understand correctly, you have a glitchy au...
whats harvest and whats dio
lol I'm very new to this sort of thing, but I'm really interested in learning it all, but for now the first step is just getting it to work in the first place
obtuse bane
here is my settings
pearl pollen
Oh I'm stupid when i asked what harvest and dio where lol, sorry
obtuse bane
no problem :D
pearl pollen
no problem :D
its still realy buggy, no more static though
obtuse bane
I'm knew to apple things so I'm testing like crazy
pearl pollen
just the words are all jibberish
pearl pollen
I'm knew to apple things so I'm testing like crazy
thats really cool, I'm new to it all in general lol
rare gobletBOT
Ayo? @pearl pollen level 2 !!!
pearl pollen
do you have anyidea how to make it so the words are more easy to understand as of now you cant make out a single thing being said
obtuse bane
do you have anyidea how to make it so the words are more easy to understand as o...
idk but I've noticed that some models are more difficult to run than others
Maybe try other models ? idk
pearl pollen
idk but I've noticed that some models are more difficult to run than others
do you have any specfic models that are easy to run that you know if
obtuse bane
I've just compared some models of my own and the defaults one (wich are really glitchy)
obtuse bane
do you have any specfic models that are easy to run that you know if
mines 💀
obtuse bane
do you have any specfic models that are easy to run that you know if
pick the one u want but I'm assuming you're not french so it might not be that flawless
also are you on an 8g or 16g machine ?
pearl pollen
also are you on an 8g or 16g machine ?
8g
obtuse bane
yeah it migh be the problem
I think 8g isn't enough
you can try higher chunk values tho
pearl pollen
i have it hooked up to a mini pc that has 10 tb of memory though
obtuse bane
so you have 10TB on you're mac rn ????
pearl pollen
so you have 10TB on you're mac rn ????
its external
obtuse bane
is it using the 10tb ?
pearl pollen
yea
obtuse bane
how did you paired it ?
pearl pollen
idk if it's a good idea or not
I just plug it into it whenever I want to play a bigger game
The rest of the time it’s just a regular Mac
pearl pollen
I have , I get even worse results but I’ll try it again anyway
obtuse bane
tf
pearl pollen
hang on im copping it to my desktop
copying*
alright im opening it now
yea its even worse then before
pearl pollen
tf
most of the time after it lags the voice speaks but then gets cut off before it can fishish speaking
rare gobletBOT
Ayo? @pearl pollen level 3 !!!
pearl pollen
here are my current settings I am having a lot of delay in the voice as well as a lot of static, most of the time the voice sounds like it is speaking gibberish. I am on a MacBook air with an M1 chip, can someone please help me, I would appreciate it very much.
knotty moth
The rest of the time it’s just a regular Mac
what? are you trying to stream your game and okada using a mac?
pearl pollen
what? are you trying to stream your game and okada using a mac?
im not trying to stream anything?
im just trying to get this thing to work
knotty moth
here are my current settings I am having a lot of delay in the voice as well as ...
is it 17b mac installation? anyway M1 is bare minimum but you may not get good enough performance
pearl pollen
is it 17b mac installation? anyway M1 is bare minimum but you may not get good e...
oh ok
thank you
pearl pollen
is it 17b mac installation? anyway M1 is bare minimum but you may not get good e...
i apricate your help
obtuse bane
np
I've got some troubles too!
First time on apple silicon and after a test run on rvc applio, I've got an error when trying to create the index file: Failed to train index: need at least one array to concatenate
I don't know if it's macos related or not... the last time I used rvc was one year ago and I saw a lot of changes
deft finch
What are the differences in all the models?
What are local forks?
Whats the difference between collabs / spaces and local forks?
Are there any other models not listed in the info page of rvc?
How many rvc versions are there?
Does each different model have different versions?
Does every rvc model use the same voice models?
Can newer versions of rvc use older voice models?
low shard
What are local forks?
Local (on your good PC) modified versions of a program, in this case RVC
deft finch
Local (on your good PC) modified versions of a program, in this case RVC
thanks
low shard
Whats the difference between collabs / spaces and local forks?
Google colabs, hugging face spaces & kaggles are on CLoud, so running remotely on a good PC with GPU, while locally means u do it on your PC without needing of using those services
low shard
How many rvc versions are there?
RVC V1 & V2, as per models as I don't think you are referring to forks, everyone uses v2, the difference should be only the pretrain though
crisp orchid
can someone help i can hear my mic on the app but it does not go through to my actual mic
low shard
What are the differences in all the models?
Like, it's models of different people
So the voice
low shard
Are there any other models not listed in the info page of rvc?
Wdym? You need ways to search rvc models?
low shard
Sorry, I was thinking of the ai models
All good
deft finch
No, I was wondering if those were all of them
low shard
No, I was wondering if those were all of them
Wdym exactly with 'those'
deft finch
Wdym exactly with 'those'
The different ais
low shard
Can newer versions of rvc use older voice models?
Yes, but using RVC V1 models is not much suggested as they are prolly lower quality
deft finch
Yes, but using RVC V1 models is not much suggested as they are prolly lower qual...
Ok thank you
low shard
The different ais
I don't understand what exact models are listed in the RVC info page
deft finch
I don't understand what exact models are listed in the RVC info page
Ok, what is the most used model of rvc?
knotty moth
How many rvc versions are there?
all custom pretrains in #1235952130855010365 use v2
low shard
Ok, what is the most used model of rvc?
I don't exactly know which is the most used, there are over thousands of rvc voice models
deft finch
I don't exactly know which is the most used, there are over thousands of rvc voi...
No, not voice model, but rvc version or model, I dont know what to call them
low shard
No, not voice model, but rvc version or model, I dont know what to call them
So you are saying like if there are other versions like RVC V3?
deft finch
So you are saying like if there are other versions like RVC V3?
No, like crepe
in the list is shows a bunch of differents types of rvc
applio rvc fork
replay
and a bunch of others
huggingface
low shard
No, like crepe
Ohh, that's the f0 method/pitch extraction algorithm, you can see them in https://docs.aihub.wtf/rvc/resources/inference-settings/#pitch-extraction-algorithm , the most used one is rmvpe as its good quality and isn't as noise sensitive as mangio crepe
deft finch
Ohh, that's the f0 method/pitch extraction algorithm, you can see them in https:...
ok thanks
low shard
in the list is shows a bunch of differents types of rvc
That's the forks, not exactly RVC architectures
There are different forks but there isn't a best one
deft finch
There are different forks but there isn't a best one
Do all the forks use rvc v1 and v2 or do some of them only use rvc v2?
low shard
Like applio is a fork of RVC, Ilaria RVC mainline is a fork of rvc
deft finch
Like applio is a fork of RVC, Ilaria RVC mainline is a fork of rvc
ok thanks
low shard
Do all the forks use rvc v1 and v2 or do some of them only use rvc v2?
That's an option that you can select when training a model in all forks
deft finch
That's an option that you can select when training a model in all forks
thank you
low shard
It's an option only for when you train (make) the model
low shard
Does every rvc model use the same voice models?
Well technically the RVC models are compatible with all RVCs (with this I include the forks), there may only be some issues when using models whose download link is from Google drive or Weights.gg when you are using a cloud RVC, but that can be fixed (for example you can download the models files and usually there's a way to manually upload them into the RVC), so if you ever run into any similar issues you can ask here
obtuse bane
I've got some troubles too!
First time on apple silicon and after a test run on ...
Anyone ?
sage wasp
hey I dont have a "Kamolaze"
buoyant fog
I cant hear myself whenever I talk with the ai could anybody help me please?
wicked shore
Why doesn’t the voice changer work for me someone plz help dm me
violet heron
Why doesn’t the voice changer work for me someone plz help dm me
!howtoask
wicked shore
Ok
So I am trying to get the vb cable installed but it says access denied “be sure to run this program in administrator mode (right click on setup program to do so) “
!howtoask
rare gobletBOT
Ayo? @wicked shore level 1 !!!
wicked shore
@violet heron
violet heron
So I am trying to get the vb cable installed but it says access denied “be sure ...
Don’t use vb audio
Use vac lite instead
Vb audio often produces glitchy and cutting when you try to use it on other apps
wicked shore
Oh ok
violet heron
Run setup64
And make sure it’s not your default mic and speakers in audio properties on windows
wicked shore
thanks
dawn crown
it only lets me on certain voices
violet heron
it only lets me on certain voices
Make sure they are RVC models
Because some models in #1175430844685484042 are GPT so Vits and not RVC
dawn crown
Make sure they are RVC models
literally ur the best bro
rare gobletBOT
Ayo? @dawn crown level 1 !!!
dawn crown
wtf
violet heron
wtf
If you can’t find the RVC model you want check if weights gg has the model
You will need a (free) account to download models off there tho
Just a little heads up is that models aren’t held up to the same standards on here so models may sound weird
dawn crown
where can i find more rvc models? i feel like where im looking is kind of limited
crisp orchid
how can i make my voice changer less raspy?
violet heron
where can i find more rvc models? i feel like where im looking is kind of limite...
as I just said weights.gg
violet heron
how can i make my voice changer less raspy?
Use RMVPE instead of crepe or harvest
And use clean vocals with no music
dawn crown
as I just said weights.gg
you can download off there?
or you mean just using it directly
violet heron
you can download off there?
If you create a account
And click the 3 dots and click download
wispy lodge
here are my current settings I am having a lot of delay in the voice as well as ...
You can try running a modified version that performs better. There are installation instructions in readme
https://github.com/deiteris/voice-changer
wispy lodge
here is my settings
You could also try it on m2 😄
fickle tartan
Can someone help me?
! C:\Users\AppData\Local\Temp\Rar$DIa13980.42270.rartemp\voice-changer-windows-amd64-cuda.zip (2).001: The file is corrupted
! Cannot execute "C:\Users\AppData\Local\Temp\Rar$EXa16076.43797.rartemp\MMVCServerSIO\MMVCServerSIO.exe"
wispy lodge
Can someone help me?
! C:\Users\AppData\Local\Temp\Rar$DIa13980.42270.rartemp\v...
Did you follow the instructions here?
https://github.com/deiteris/voice-changer?tab=readme-ov-file#for-nvidia-users
fickle tartan
Did you follow the instructions here?
https://github.com/deiteris/voice-changer?...
I just solved it, I had both in a winrar and it had to be in a folder
knotty moth
Can someone help me?
! C:\Users\AppData\Local\Temp\Rar$DIa13980.42270.rartemp\v...
try redownloading and make sure there's no antivirus interfering
fickle tartan
Did you follow the instructions here?
https://github.com/deiteris/voice-changer?...
After running it and installing, what do I have to do next?
knotty moth
After running it and installing, what do I have to do next?
read the guide https://rentry.co/VoiceChangerGuide
fickle tartan
read the guide https://rentry.co/VoiceChangerGuide
I already did that but I get this error.
Traceback (most recent call last):
File "client.py", line 17, in <module>
File "asyncio\runners.py", line 190, in run
File "asyncio\runners.py", line 118, in run
File "asyncio\base_events.py", line 654, in run_until_complete
File "main.py", line 114, in main
File "downloader\WeightDownloader.py", line 87, in downloadWeight
Exceptions.WeightDownloadException: 'Failed to download weight.'
wispy lodge
I already did that but I get this error.
Traceback (most recent call last):
Fil...
Does it show which files were verified or downloaded above the error?
If some of them fail to download, you can download missing files and put them into pretrain folder separately
https://drive.google.com/drive/folders/1OFfM9rmxCZkiYjxoK_yzhRbcXpt0TiJ0
fickle tartan
Does it show which files were verified or downloaded above the error?
Only shows this
Python: 3.11.9 (tags/v3.11.9:de54cf5, Apr 2 2024, 10:12:12) [MSC v.1938 64 bit (AMD64)]
Voice changer version: b2154 NVIDIA-CUDA
Activating the Voice Changer.
[Voice Changer] Loading weights.
[Voice Changer]
[Voice Changer]
[Voice Changer]
[Voice Changer]
[Voice Changer]
Traceback (most recent call last):
File "client.py", line 17, in <module>
File "asyncio\runners.py", line 190, in run
File "asyncio\runners.py", line 118, in run
File "asyncio\base_events.py", line 654, in run_until_complete
File "main.py", line 114, in main
File "downloader\WeightDownloader.py", line 87, in downloadWeight
Exceptions.WeightDownloadException: 'Failed to download weight.'
wispy lodge
Try restarting it then, it should show the files that were downloaded
fathom reef
So, what kind of information needs to match between two different voice models to merge them together?
Because I have a character which is a higher-pitched voice of myself, and I wanted to merge his voice with another character to give him a voice that somewhat matches what his kind actually sounds like.
But I can't do it because the two voice models "don't match."
Both of their information match, yet it still outputs the error.
Do the epochs also need to be the same?
Because I can't do that. It would take too long for me to train a whole new model just for it to be scrapped out of existence.
wispy lodge
Only sample rate must match
fathom reef
They do, yet Ilaria is still not behaving.
knotty moth
So, what kind of information needs to match between two different voice models t...
you should only do merge two/few checkpoints from the same model dataset and configurations, read the following guide
fathom reef
you should only do merge two/few checkpoints from the same model dataset and con...
Okay, I do this just as a hobby. I do not understand a single thing you're saying.
knotty moth
Okay, I do this just as a hobby. I do not understand a single thing you're sayin...
yeah unless you know how to train models
fathom reef
For context, these are the two models. I'd rather not reveal the second character, though. I'm just not ready to reveal the full final character yet.
fathom reef
yeah unless you know how to train models
I only used RVC Disconnected to train my character.
Don't ask why my character is only 50 epochs. It's because I didn't have much voice lines for him.
fathom reef
yeah unless you know how to train models
But yeah, Ilaria won't let me merge the two models, regardless of the two models having the same information.
knotty moth
But yeah, Ilaria won't let me merge the two models, regardless of the two models...
have you tried on local mainline RVC?
fathom reef
have you tried on local mainline RVC?
I don't have a good enough graphics card to do so.
(aka AMD Radeon RX 580)
knotty moth
I don't have a good enough graphics card to do so.
I dont think it needs so
wispy lodge
You don't even need a gpu to merge. It's done on cpu easily
fathom reef
Okay, so I've never used local RVC stuff before. I've decided to use the Manigo fork, but I have no idea which file to run first.
fickle tartan
Try restarting it then, it should show the files that were downloaded
These files were missing right when the error occurred, but when I try to put the files I get this another error:
! C:Users####OneDriveDesktopasdsadadsvoice-changer-windows-amd64-cuda.zip.001: The file is corrupted
rare gobletBOT
Ayo? @fickle tartan level 2 !!!
fickle tartan
I already tried downloading it again and disabling the antivirus but it gives the same error
knotty moth
These files were missing right when the error occurred, but when I try to put th...
try in the root drive C: or D:, not fucking onedrive
thin stump
Okay, I do this just as a hobby. I do not understand a single thing you're sayin...
lmao this is a hobby including the complicated stuff like that graph
wispy lodge
These files were missing right when the error occurred, but when I try to put th...
Are you trying to put the files inside the zip file?... And yeah, don't work inside OneDrive folder, move it somewhere else
fathom reef
lmao this is a hobby including the complicated stuff like that graph
That's not what I meant at all.
I meant I just do the basic stuff.
I am not an AI program developer. At all.
I just use the programs. That's all I do.
I just want to do fun stuff with AI. I don't care about this whole graph stuff.
fickle tartan
try in the root drive C: or D:, not fucking onedrive <:nails:1159569314848972891...
! C:\asdsadads\voice-changer-windows-amd64-cuda.zip.001: The file is corrupted
wispy lodge
Do you have 002 in the same folder?
fickle tartan
Do you have 002 in the same folder?
Yes
carmine hearth
For context, these are the two models. I'd rather not reveal the second characte...
I'm using machine translation, so I apologize if I'm not making sense.
Is your model not working correctly with inference? Did you choose to train with something other than contentvec? Did you put spaces/special characters in your model name?
thin stump
I'm using machine translation, so I apologize if I'm not making sense.
Is your m...
He is trying to merge models. That is more advanced for the average person to understand especially when you do not want to read any guides
fathom reef
The problem is that it doesn't generate any indexes. Therefore, any inferences on Ilaria cannot be done.
knotty moth
The problem is that it doesn't generate any indexes. Therefore, any inferences o...
the same model dataset and configurations also produce and use the same index file, wdym
fathom reef
I never got any indexes out of the merged model from Mangio.
knotty moth
I never got any indexes out of the merged model from Mangio.
yeah the index file can only be created through index training as well as preprocessed files
fathom reef
yeah the index file can only be created through index training as well as prepro...
Yeah, so I don't think Ilaria accounted for that...
torpid oriole
hey guys i have a problem please give me a solution i trained a model which was completely good and i tried with my own voice i recorded a audio note and changed it using that model and it was perfect butt when i use owkada for live conversion it sounds wired i shared that model to others and they used that in live conversion its perfect but when im using that its like someone just wanna close my mouth using cloth like that in every female model when i use in live conversion its like something on my mouth but when i record it and convert it , its perfect . but same time when others try that same model it was perfect for them in live conversion ..... please give me a solution ....
late crane
[Voice Changer] wait for loading model... -1 -1
[Voice Changer] [Exception] 'No model for model_common loaded. Please confirm the model uploaded.'
obtuse bane
You could also try it on m2 😄
I'm on m2....
late crane
what's the problem
wispy lodge
I'm on m2....
Well, and I'm saying that you can also try the modified verison
obtuse bane
My bad my connexion didn't let me see the embed above your reply so I thought you asked me to give you some settings for m2
Btw any idea ?
First time on apple silicon and after a test run on rvc applio, I've got an error when trying to create the index file: Failed to train index: need at least one array to concatenate
I don't know if it's macos related or not... the last time I used rvc was one year ago and I saw a lot of changes so idk if there is a specific version for macos
wispy lodge
My bad my connexion didn't let me see the embed above your reply so I thought y...
Ah, no. I was just saying that there's a modified version that was tested with macOS arm recently and it has actually good results even on m1
Like 256ms delay with fcpe and 131k extra
obtuse bane
I'll give it a try
past trench
I need best web rvc colab to make train
rare gobletBOT
Ayo? @past trench level 2 !!!
carmine hearth
-colab
carmine hearth
Applio, RVC Disconnected, RVC Mainline
thin stump
I need best web rvc colab to make train
-kaggle
thin stump
the guide was recently finished
thin stump
I can just go eep while it trains
. So good
low shard
the voice changer accent sometimes changes
this is the help channel for rvc for ai covers, not wokada for calls, ask in #1192011222023950368 & #🔍│help-w-okada
Also i can't help you
I don't use wokada locally so you need to wait for someone in those channels to help yu
proven hill
it doesnt follow my accent sometimes
set the index to 0
glass kayak
set the index to 0
ok
gray tinsel
what headset is the highest possible quality for rvc models?
budget is whatever, i need replacement eventually anyways.
knotty moth
what headset is the highest possible quality for rvc models?
budget is whatever,...
you mean "bare minimum", which are any decent cheap two-digit buck ones?
gray tinsel
you mean "bare minimum", which are any decent cheap two-digit buck ones?
why go "bare minimum"? trying to improve hardware and voice accuracy much as possible
🤔 something that'll pick up my voice with no faults
knotty moth
why go "bare minimum"? trying to improve hardware and voice accuracy much as pos...
even buzzing noises in cheap mics are quiet enough to handle by any noise suppression including the voice changer's built-in
so you don't really need studio-grade/streaming dedicated mics or any snake oil stuffs
slim berry
can someone help me i didn't get how to use trained voice
gray tinsel
even buzzing noises in cheap mics are quiet enough to handle by any noise suppre...
oh! i got a few of those headsets local, next time i'll grab one
proper shale
can someone help me i didn't get how to use trained voice
https://docs.aihub.wtf/ - how to make AI cover guide should do it
or use ilaria rvc
https://docs.google.com/document/d/1YbXcLFPaGjhOdG5NFkK3QrucCEpHZBwFUxkeMO8aB18/edit
proven hill
i reccomend using ilaria rvc because yea
proper shale
i reccomend using ilaria rvc because yea
- totally not Ilaria (the creator of Ilaria RVC)
gray tinsel
this tech is fascinating so all the way, can't wait to escape gtx1070
proven hill
- totally not Ilaria (the creator of Ilaria RVC)
self promo is allowed for me 
past trench
Please help me I want to refresh my uncles voice just make it a zip model for me , Anyone who helps me should be DM
proper shale
Please help me I want to refresh my uncles voice just make it a zip model for me...
just #1159289738314919936 or learn how to make it in https://docs.aihub.wtf/
mental wagon
After training a model in the mainline webui, where do i find the path & index files?
proper shale
After training a model in the mainline webui, where do i find the path & index f...
pth on weights folder, index on logs/your model's folder
past trench
just <#1159289738314919936> or learn how to make it in https://docs.aihub.wtf/
When it's a help room, no one comes to help anyone
proper shale
When it's a help room, no one comes to help anyone
well I'm trying my best to help you man
- i gave you the way to either ask someone to do it for you or make it yourself
past trench
well I'm trying my best to help you man
I really love helpful and initiative men 🖤
ionic geyser
Guys as soon as I make it high rvc in my voice changer client it stops the audio and stops me from hearing myself?
proper shale
I really love helpful and initiative men 🖤
proper shale
Guys as soon as I make it high rvc in my voice changer client it stops the audio...
you put RVC quality as high? not worth it
ionic geyser
Why not?
proper shale
<@622856015444049937> do you use this one too
yeah
proper shale
just keep as low
past trench
+ i gave you the way to either ask someone to do it for you or make it yourself
Well, I want to try myself, but my laptop will be running Cora i3 and Windows 7
proper shale
Well, I want to try myself, but my laptop will be running Cora i3 and Windows 7
you can just use Colab
https://docs.aihub.wtf/rvc/cloud/rvc-disconnected/
rvc disconnected
past trench
you can just use Colab
I used easygui colab but the index and the download model gave an error as to whether it was my laptop error or what
rare gobletBOT
Ayo? @past trench level 3 !!!
proper shale
I used easygui colab but the index and the download model gave an error as to wh...
probably screwed up on something, and no it ain't your computer's fault
past trench
probably screwed up on something, and no it ain't your computer's fault
There is currently a problem-free colab bro
thin stump
There is currently a problem-free colab bro
Have you read the mainline guide I sent you 🤔. You literally cannot get lost unless you are babuu
past trench
Have you read the mainline guide I sent you 🤔. You literally cannot get lost un...
No, I can't read all that, but please tell me which collab is perfect bro 🖤
proper shale
No, I can't read all that, but please tell me which collab is perfect bro 🖤
that's the guide for the colab
you must read it
past trench
that's the guide for the colab
I did the go web and this one comes up
proper shale
I did the go web and this one comes up
...you shouldnt download the actual mainline mate
just use the colab
past trench
Well, brother, don't find my colab
violet heron
Well, brother, don't find my colab
-colab
thin stump
The guide is done by the way. I went to sleep for most of this usage quota and it's very convenient
@proven hill I'm definitely doing something wrong here
I trained straight out of mainline with those pth files and now I want to merge them
proven hill
<@740255187234717838> I'm definitely doing something wrong here
- my name is I L A R I A only one l, Ilaria not Illaria lol
- the error is in the name, it cant find the first pth
thin stump
doh, ok thanks
proven hill
np
elfin sun
Hello what is passthru?
Also i dont think any voice model is working for me, My original voice is playing as it is
knotty moth
Hello what is passthru?
Also i dont think any voice model is working for me, My...
if the passthru button is red, toggle it
elfin sun
VB audio virtual cable is only getting the input if passthru is on (red)
rare gobletBOT
Ayo? @elfin sun level 1 !!!
elfin sun
For some reason, after adding the virtual cable, this keeps on going, and the voice changer doesn't load, can someone help?
fiery raptor
why does this stay blank even tho there is clearly a file here? (iralia mainline rvc btw)
fossil condor
-local
viral folio
what is the link for the colab to make an ai cover_
violet heron
what is the link for the colab to make an ai cover_
Ilaria RVC zero is recommended over using colabs
bold lark
-local
hard gyro
hello guys help me please i need best version for my rtx3070ti and ryzen 7 7700
glad veldt
what's the best background vocal remover?
soft stratus
I had a problem with the boy's voice as a female actor... Do you know what's problem
violet heron
I had a problem with the boy's voice as a female actor... Do you know what's pro...
Change pitch
open flint
can someone help me
soft stratus
Change pitch
I tried but i want arabian guy to help me
violet heron
can someone help me
Don’t ask to ask, just ask
upper copper
Dude colab keeps giving an error, what should I do
open flint
Don’t ask to ask, just ask
huh?
rare gobletBOT
Ayo? @open flint level 1 !!!
brave garnetBOT
Made with love by Ilaria 💖
⠀
Settings for Nvidia GPUs 
F0 Det.: rmvpe (suggested for all series)
RTX 40-series: 80-96 chunk | +16384 extra
RTX 30-series: 96-112 chunk | +16384 extra
RTX 20-series: 112-128 chunk | +16384 extra
GTX 16-series: 128-192 chunk | +8192 extra
GTX 10-series: 128-192 chunk | +8192 extra
Advanced Settings
Protocol : Sio or Rest
Crossfade: 4096 start 0.2 end 0.8
Trancate: 300
Silencefront: Off
Protect: 0.5
RVC Quality: Low
⠀
Made with love by Ilaria 💖
fiery raptor
why does this stay blank even tho there is clearly a file here? (iralia mainline...
can somebody tell me why this is happening btw
thin stump
can somebody tell me why this is happening btw
Do the inferencing here https://docs.google.com/document/d/1YbXcLFPaGjhOdG5NFkK3QrucCEpHZBwFUxkeMO8aB18/edit
fiery raptor
Do the inferencing here https://docs.google.com/document/d/1YbXcLFPaGjhOdG5NFkK3...
i forgot how slow the huggingface rvc gets but thanks!
brittle wing
-colab
rare gobletBOT
Ayo? @brittle wing level 1 !!!
distant citrus
Hi! Im new into this of AI voice mod, and im trying help a friend with a mod ||Not gonna show more details|| and we're trying to make an audio, but the sound effect in the back, sounds the same as the original audio, any idea how to make it also sound as the ai voice?
crude bolt
Hi! Im new into this of AI voice mod, and im trying help a friend with a mod ||N...
morning
to understand what you want , we first need to align in the terminology. It might be easier if we just have a talk (on the voicechannels). However that's up to you. I am not sure what you are trying to achieve.
knotty moth
Hi! Im new into this of AI voice mod, and im trying help a friend with a mod ||N...
wdym "mod"? if you want to change the voice by inferring it, first you have to separate it from other background music/sounds using an application like https://huggingface.co/spaces/TheStinger/UVR5_UI
then you can infer it using RVC application like https://docs.google.com/document/d/1YbXcLFPaGjhOdG5NFkK3QrucCEpHZBwFUxkeMO8aB18
also read the guide including other RVC solutions https://docs.aihub.wtf/
distant citrus
to understand what you want , we first need to align in the terminology. It migh...
Sorry notification didnt sound.
So, im trying to make a voice for the sounds of the box of Black Ops Zombies, on this voice theres a reverb on the bacground that repets what the main voice said, the main voice indeed change, but the backgrounds voices no
Sorry for bad english.
crude bolt
you need some after effects plugins to chain on the rvc program , there are no effects included.
knotty moth
try this fork for better performance and download the dml one for your AMD gpu https://github.com/deiteris/voice-changer
knotty moth
I don't think it's necessary for the usual precompiled installation
minor linden
can someone help me set up the rvc i downloaded it but dont know how to run it
rare gobletBOT
Ayo? @grand mauve level 1 !!!