no its all good ! we spoke like 3 times anyway so i assume you had forgotten me anyway haha
#✨│ai-help
devout field
devout field
im trying this right now
proven hill
valid help, follow it
rare gobletBOT
Ayo? @devout field level 4 !!! 
devout field
which one
real field
gabru bro does msvc143 work because i cant find 142
search "MSVC V142"
real field
the most recently (10.0.20348.0)
devout field
okay thank you!!
real field
if possible, install it too
i guess that english its like "Desktop Development with C++"
real field
u re welcomee
rare gobletBOT
Ayo? @real field level 2 !!!
devout field
tts
applio so far seems to have the best tts software
thats like the least robotic
proven hill
ilaria rvc use the same tts
devout field
i tried yours ages ago couldnt make it work
proven hill
devout field
and ilaria not to mention i wanna be able to run it locally
violet heron
applio so far seems to have the best tts software
Applios TTS is just edgetts inferred over
So you can just use mainline and convert edgetts results with mainline
proven hill
or use ilaria rvc mainline but its EOL (like applio)
proven hill
why locally tho when you can simply use it with 1 click
old oracle
devout field
why locally tho when you can simply use it with 1 click
bc my pc is super fast
env\python.exe: can't open file 'C:\Windows\system32\app.py': [Errno 2] No such file or directory
Press any key to continue . . .
also this line is what seems to be the error
proven hill
bc my pc is super fast
you need training or not
violet heron
ive installed the zip file
Did you unzip it
devout field
yes yes
radiant juniper
how do i make a minecraft villager song?
violet heron
how do i make a minecraft villager song?
do you want to do it locally using your computers hardware or using a website
violet heron
Or like an ai cover that they can hear
Yeah which one do you want
Covers or realtime
radiant juniper
Real-time
rare gobletBOT
Ayo? @radiant juniper level 1 !!!
radiant juniper
Intel
proven hill
rip
radiant juniper
Is it not possible?
edgy ember
where to download rvc
radiant juniper
Both
pastel oak
Is it not possible?
If you mean an Intel HD Graphics, its too weak to run a realtime voice changer
If you have a mp3 file ready that you want to play in voice chat then you dont need this, you could install something like a soundboard and play it in voice chat? Is that something you want instead? Sounded like it at first before you mentioned realtime
violet heron
It says uhd
Is there a GPU 1?
radiant juniper
Is there a GPU 1?
Nope
radiant juniper
radiant juniper
If you mean an Intel HD Graphics, its too weak to run a realtime voice changer
...
But if this is an option I could use it but I’d need a free soundboard if you provide any
pastel oak
timbo do you know any soundboards to run with virtual cable? lol i dont know any
not AI related
Else check out any yt tutorial regarding a "soundboard to use in voice chat". Idk more than that :/
devout field
Did you unzip it
i have extracted it to my ssd and still get the same error, have u got time to help me out
violet heron
i have extracted it to my ssd and still get the same error, have u got time to h...
Why not use mainline?
devout field
is that within the link u sent me?
devout field
Why not use mainline?
i got the last one 😭
pastel oak
is that within the link u sent me?
https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/releases/tag/2.2.231006
download whichever applies to ur gpu, nvidia or amd/intel
this is Mainline
night rune
Can someone suggest what caused this to happen? And is there any way I can continue model training without starting over (I have a backup)?
pastel oak
Can someone suggest what caused this to happen? And is there any way I can conti...
I dont know what caused it but to answer the 2nd question, save the D and G file on the 2nd Screenshot, with those you can continue training
devout field
https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/releases/t...
thank you
night rune
I dont know what caused it but to answer the 2nd question, save the D and G file...
Should they be moved to the same path where they were? And then what do I do after that? Just click “Train model”?
pastel oak
Should they be moved to the same path where they were? And then what do I do aft...
They can technically be anywhere but no harm in keeping them in the same logs folder, probably even recommended.
Choose the same model name, then on the last step select those d and g paths where it asks for the D and G path, then you can train. On the commandline youll see "loaded ... (epoch X)" number, ai think
night rune
They can technically be anywhere but no harm in keeping them in the same logs fo...
Didn't quite understand about the paths for G and D. I use Ilaria-Mainline and this is the interface here
pastel oak
Didn't quite understand about the paths for G and D. I use Ilaria-Mainline and t...
Can you open advanced settings?
proven hill
Can you open advanced settings?
its in sample rate and pretrain
pastel oak
its in sample rate and pretrain
^ @night rune
night rune
its in sample rate and pretrain
??
night rune
what pretrain do you need
I didn't understand the question. I have the files from last training (screenshot above) and I want to continue training the same model
proven hill
I didn't understand the question. I have the files from last training (screensho...
oh i thought you needed a pretrain, sorry.
night rune
Damn translator
proven hill
you just input the same dataset and the same settings as the previous train
radiant juniper
Guys do you provide any soundboards for free?
pastel oak
Guys do you provide any soundboards for free?
If you have all the audio files on your pc you can check this out
https://youtu.be/GC1aLL7cPY4?si=6gxTHvm0Ve9VmE1x
it shows how to make the mp3 file youre playing go in as a microphone. Its not a soundboard, you have to start the mp3 file you want to play but maybe its an ok-ish solution until you find a soundboard
if i remember the name of the soundboard i used years ago ill let u know lol
tough hearth
How would i do a parody of a song using a voice model, rather than a cover of an existing song. i have the lyrics i want the parody to be.
night rune
I dont know what caused it but to answer the 2nd question, save the D and G file...
It started from the beginning 
pastel oak
It started from the beginning <:boohooh:1176674698629750975>
What were the options on advanced settings?
Oh no, it might have resetted your G and D paths now though since you overwritten the files
night rune
Oh no, it might have resetted your G and D paths now though since you overwritte...
You don't have to worry about that, I've got a zip file downloaded.
pastel oak
How would i do a parody of a song using a voice model, rather than a cover of an...
You can either use TTS (text to speech) but not sure if itd feel "realistic", im sure theres some advanced stuff on making a models sing with tts but would wait on someone else with more knowledge to comment if this advanced thing is of interest for you
Else you could sing it yourself and infer your vocals with the voice model?
night rune
What were the options on advanced settings?
pastel oak
No idea. Ilaria knows best, so would wait for an answer
night rune
No idea. Ilaria knows best, so would wait for an answer
Okay. Then another question, can I use something of what I have saved?
pastel oak
Okay. Then another question, can I use something of what I have saved?
Check out the weights folder for the pth file
or assets -> weights
night rune
Check out the weights folder for the pth file
There's nothing there 🥲
night rune
No idea. Ilaria knows best, so would wait for an answer
By the way, does Ilaria even show up here?
rare gobletBOT
Ayo? @night rune level 3 !!!
pastel oak
By the way, does Ilaria even show up here?
Ilaria replied to you just a few messages above. But im guessing we did something wrong or Ilaria misunderstood your question
night rune
Ilaria replied to you just a few messages above. But im guessing we did somethin...
I guess I'm blind, hahaha
night rune
*maybe* you can just load up the G file? Never tried
Xd
night rune
What gpu does your pc have btw
I'm using colab
pastel oak
Asking in case you got an nvidia you could quickly train one epoch to get the file 😂 idk if thats efficient
rare gobletBOT
Ayo? @pastel oak level 53 !!!
night rune
I'm using colab
cuz my gpu is gtx 1650 super
pastel oak
Probably would work haha
Or you get "rvc disconnected" colab, it should have the resume train feature to train 1 epoch to test it out
night rune
Screw it, I'm going to bed. Thanks for trying to help.
pastel oak
Yee tomorrows a better day
vale osprey
-colab
brittle wing
Heya, I tested Kaggle and it was top-notched, I didn’t know it trained so fast than Colab! Thank you for showing me!
By the way, the limit of training epochs is 1000, but is there any way I can go over 1000?
-kaggle
odd shale
Heya, I tested Kaggle and it was top-notched, I didn’t know it trained so fast t...
Just put a number that goes beyond 1000 epochs if you want when training
But you'll risk getting your model overtrained unless you check tensorboard
jolly onyx
-colab
knotty moth
Heya, I tested Kaggle and it was top-notched, I didn’t know it trained so fast t...
check tensorboard before deciding to go beyond 1000, or you can have me send you a modified runmain.py (the main gui)
brittle wing
But you'll risk getting your model overtrained unless you check tensorboard
Yeah, I'll try and make sure to look at tensorboard, finding the lowest point too!
spark mason
Need someone who can AI my voice for me. you will get paid!! $ 💰 Dm me!!
crude flame
Need someone who can AI my voice for me. you will get paid!! $ 💰 Dm me!!
Go to #1191429836321849435 or make a paid request in #1159289738314919936
spark mason
thank u so much gng
teal ridge
Somehow when I submitted my "Goblin Caught on Tape" model to this server, I noticed it didn't convert to weights.gg.
Not only that, it seems I have no access to sharing models to weights.gg no more because of the same fucking "Failed to add model: Network Error." Tried it on Google Chrome and Microsoft Edge.
thin stump
@willow lichen
thorny sail
guys, i have a problem, i have rx5700xt but voice changer use cpu, i try all gpu setting in realtime voice changer, maybe you have decision?
snow jolt
what is the best pretrain for training a model
pastel oak
guys, i have a problem, i have rx5700xt but voice changer use cpu, i try all gp...
Download this, its optimized for AMD gpus
brittle wing
-local
brittle wing
Which Local has more recent or gets recent updates?
thorny sail
Download this, its optimized for AMD gpus
https://rentry.co/ForkVoiceChangerGui...
ok, i try it
brittle wing
Download this, its optimized for AMD gpus
https://rentry.co/ForkVoiceChangerGui...
Does AMD drivers work with any kind of Ai Training? 7900 xtx?
wispy lodge
Does AMD drivers work with any kind of Ai Training? 7900 xtx?
I think you've already asked and at the moment the answer is no. There's pending Zluda support in Applio that would allow training with AMD GPUs
https://github.com/IAHispano/Applio/pull/513
brittle wing
I think you've already asked and at the moment the answer is no. There's pending...
Still no? dang. Ill check Zluda thank you
knotty moth
does anybody know why easyGUI rvc isn't working its like
it won't be fixed anymore tho
distant hamlet
does anybody know why easyGUI rvc isn't working its like
!pip install pip==24.0
add to the very top of the notebook and run it before anything else.
brave bane
`!pip install pip==24.0`
add to the very top of the notebook and run it before a...
did you also have this error??
distant hamlet
yes. bc my other Colab notebooks broke the other day too. this fix has so far been working for me on multiple notebooks, including easyGUI
tried it again this morning too bc someone else asked
brave bane
oh
ok
well all its doing now is making the first cell run slower but thas fine
nvm
it just finished
ok
time to see if it works
oh
rare gobletBOT
Ayo? @brave bane level 2 !!!
distant hamlet
yeee
rare gobletBOT
Ayo? @distant hamlet level 1 !!!
brave bane
how should 3 min .mp3 take to upload, i've only tried a buncha short ones not full songs.
its been like 10 minutes and still not done
two 3 min
distant hamlet
how should 3 min .mp3 take to upload, i've only tried a buncha short ones not fu...
uploading/downloading directly on Colab is always extra slow, but for 3 min mp3 shouldn't take v long.. personally i normally like to upload on my Drive first tho, then mount and drag it over in Colab
brave bane
yeah uploaded the files directy to the public url
distant hamlet
what's the upload progress now? lol
distant hamlet
weird bc those files are tiny. but try other method next time, just have to snag the Drive mount code for easyGUI
thorny sail
Download this, its optimized for AMD gpus
https://rentry.co/ForkVoiceChangerGui...
after start i have console spam, i try to load other model and change settings, nothing change Text:onnxruntime.capi.onnxruntime_pybind11_state.InvalidArgument: [ONNXRuntimeError] : 2 : INVALID_ARGUMENT : Invalid input name: skip_head
2024-08-01 13:50:54,295 ERROR [VoiceChangerManager] [ONNXRuntimeError] : 2 : INVALID_ARGUMENT : Invalid input name: skip_head
Traceback (most recent call last):
File "voice_changer\VoiceChangerManager.py", line 222, in changeVoice
audio, vol, perf = self.voiceChanger.on_request(receivedData)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "torch\utils_contextlib.py", line 115, in decorate_context
return func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "voice_changer\VoiceChangerV2.py", line 137, in on_request
audio, vol = self.voiceChangerModel.inference(audio_in)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "voice_changer\RVC\RVCr2.py", line 246, in inference
audio_model = self.pipeline.exec(
^^^^^^^^^^^^^^^^^^^
File "voice_changer\RVC\pipeline\Pipeline.py", line 234, in exec
out_audio = self.inferencer.infer(feats, p_len, pitch, pitchf, sid, skip_head, return_length, formant_length).float()
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "voice_changer\RVC\inferencer\OnnxRVCInferencer.py", line 64, in infer
output = self.model.run(
^^^^^^^^^^^^^^^
File "onnxruntime\capi\onnxruntime_inference_collection.py", line 220, in run
return self._sess.run(output_names, input_feed, run_options)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
rare gobletBOT
Ayo? @thorny sail level 1 !!!
pastel oak
after start i have console spam, i try to load other model and change settings, ...
@wispy lodge Could you have a look at this 🙏 gpu is rx5700xt if it matters
wispy lodge
after start i have console spam, i try to load other model and change settings, ...
Upload original pth model, then export it to onnx if necessary
thorny sail
look at program?
wispy lodge
Onnx models exported in original version don't work
thorny sail
Onnx models exported in original version don't work
i try it
tawny needle
Can I use ASIO driver with RVC Web UI?
tawny needle
real ASIO driver from audio interface
tawny needle
not ASIO wrapper like flexasio, asio4all
next geyser
though i can hear my voicce
pastel oak
voice changer is not outputting any audio
If your passthru is blinking red, then click on it to turn it off so its green
tawny needle
I couldn't select ASIO input, output device with w-okada
wispy lodge
I couldn't select ASIO input, output device with w-okada
Which version?
wispy lodge
Like, ASIO device does not appear or there's an error? 🤔
tawny needle
ASIO driver appears in list but I can't select ASIO devices
wispy lodge
i try it and it works but with 2000ms delay, it dont use resourses
Could you show your settings?
thorny sail
Could you show your settings?
oh, i use the wrong gpu, i have 2 same names
thanks you for help
tawny needle
Like, ASIO device does not appear or there's an error? 🤔
Input : only [ASIO] Audient ASIO Driver
Output : only [ASIO] Audient ASIO Driver
I can't see my mic devices
so I can't select audio routing direction
thorny sail
thanks you for help<:kittyblush:1167393107118149642>
its not working again, btw im going to play tennis with 40 pounds dumbell
tawny needle
Like, ASIO device does not appear or there's an error? 🤔
https://github.com/w-okada/voice-changer/issues/745
It's exactly the same issue as me
wispy lodge
https://github.com/w-okada/voice-changer/issues/745
It's exactly the same issue...
Huh... I'll try to check this, but I don't have a true ASIO device
tawny needle
Thank you! If you don't mind, I can support cheap audio interface for you.
ASIO audio interfaces don't work in w-okada as of now
pastel oak
its not working again, btw im going to play tennis with 40 pounds dumbell
Can you explain what doesnt work this time? Command window error log or screenshot
merry eagle
bro i am getting this error when i press download button @steel forge @urban lagoon @brittle wing @crude bolt @hearty idol
pastel oak
first dont randomly ping people, be patient and wait
secondly pinging someone in an edited message doesnt ping them
xd
merry eagle
first dont randomly ping people, be patient and wait
secondly pinging someone in...
oh thanks
pastel oak
Can you start with explaining what UI youre using and if everything else up until download worked or not? and are you using that weird button on the bottom left "download audio from this page" or are you using the inbuilt download
merry eagle
Can you start with explaining what UI youre using and if everything else up unti...
I am using apolio ui
i am pressing the original button download
merry eagle
Can you start with explaining what UI youre using and if everything else up unti...
last time i used this Ui this was perfectly working
merry eagle
Can you start with explaining what UI youre using and if everything else up unti...
2 hours ago last time i used this
pastel oak
Oke well im not sure what the issue is I dont use applio, id wait for someone else to help
merry eagle
Oke well im not sure what the issue is I dont use applio, id wait for someone el...
thats why i was tagging everyone
Last tiem there was a helper who helped me he was legit @nocturne mural
steel forge
bro i am getting this error when i press download button <@159310300275802112> <...
Dont ping random people
All your completed inferences should go to the Applio-3.2.0\assets\audios folder. The error you're getting makes me think applio was killed or something
merry eagle
Dont ping random people
All your completed inferences should go to the `Applio...
i am tagging random people . TheY all are helpers
There is no folder named applio
lilac gust
why does this happen
tawny needle
i am tagging random people . TheY all are helpers
There is no folder named appli...
tagged helpers must be annoyed
don't tag them at once
vestal vale
im using rvc and when i click on roblox the rvc wont run
like i must click on the rvc gui for the voice
if i click on the roblox window the rvc cant converse my voice
merry eagle
tagged helpers must be annoyed
i tagged them only once
pallid thicket
tame mica
tawny needle
Huh... I'll try to check this, but I don't have a true ASIO device
possible solution : https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/issues/181
fierce rock
Guys, why do I have such a lag in the discord of the voices that I add
proper shale
Guys, why do I have such a lag in the discord of the voices that I add
you're probably on the AMD version of okada, so you do need to convert downloaded models to onnx. linked below is part of the guide that will help you do that.
https://rentry.co/VoiceChangerGuide#uploading-models-amdintel-gpus-and-fixing-its-laggy-issue
pastel oak
Guys, why do I have such a lag in the discord of the voices that I add
Whats your GPU name?
proven hill
Whats your GPU name?
Mike
fierce rock
Why don't I have a graphics card in my cpu selection?
rare gobletBOT
Ayo? @fierce rock level 1 !!!
proper shale
Why don't I have a graphics card in my cpu selection?
you got the wrong version for your GPU
read this
fierce rock
I have amd
proper shale
I have amd
yeah, you probably have the Nvidia version of Okada
proper shale
MMVCServerSIO_win_onnxgpu-cuda_v.1.5.3.18a
that's for nvidia
proper shale
https://rentry.co/forkvoicechangerguide
id suggest reading the guide i linked, its the better version for amd users
fierce rock
proper shale
nothing wrong there, just open http://127.0.0.1:18888/ on your browser
fierce rock
fierce rock
An error occurred during voice conversion. Check command line window for more details.
wispy lodge
This can happen if it still plays sound after you click Stop. It still should work, but close and open voice changer again just in case
I see... I guess I could add channel selection, if it's the only thing required
tawny needle
I sincerely hope that channel selection will be added🥹
tawny needle
violet heron
royal hatch
i cant hear anything at all and this message appears : エラーが頻発しています。対象としているフレームワークのモデルがロードされているか確認してください
idk what to do honestly, i spent actual hours installing pytorch and all the rest
the stock model didn't work too
even switched versions of the voice changer and still did nothing
im talking about the one in this video
i've got it to work a long time ago, but now that im reinstalling it, the new version looks different and i cant get it to work apparently
violet heron
idk what to do honestly, i spent actual hours installing pytorch and all the res...
You don’t need PyTorch
It comes with the voice changer
Whoever said that is spreading misinfo
What’s your GPU?
royal hatch
yeah i dont remember doing that the first time, i use a 3050
rare gobletBOT
Ayo? @royal hatch level 1 !!!
violet heron
i cant hear anything at all and this message appears : エラーが頻発しています。対象としているフレームワー...
Try restarting it
Looks like it didn’t launch the base model
royal hatch
heck i reinstalled it multiple times in two versions
of wich, one had the free models and one didn't
and those didnt work too
but yeah this one doesn't have any models
its the 1.5.2.9e
violet heron
heck i reinstalled it multiple times in two versions
The Cuda or DirectML version
Which file are you opening?
royal hatch
it says cuda on the zip
violet heron
its just start_http
File extensions are prob off but it doesn’t really matter
https://github.com/deiteris/voice-changer see if this fork works maybe
royal hatch
[Voice Changer] [Exception] 'No model for model_common loaded. Please confirm the model uploaded.'
this is probably the issue
it keeps spamming even if i reload or change the model
dire wagon
name 'save_only_latest_ckpt' is not defined any fix ?
pastel oak
name 'save_only_latest_ckpt' is not defined any fix ?
Are you using webui? click on yes or no on that setting
tall rapids
Hi, I'm using RVC-GUI, I read that Crepe is the best one but I'm kind of confused by the arguments
Which would you say are the best arguments/settings
pastel oak
RVC v2 Disconnected
Search for save_only_latest_ckpt and give it a value, like "yes" or "no", thats all
tall rapids
Which would you say are the best arguments/settings
Since I can't upload a picture
Crepe hop: 64~512
Pitch: -20 ~ 20
Feature retrieval rate: 0 ~ 1
I know what pitch does, but I'm not really sure what to do with crepe hop and FRR.
I think that lower crepe hop sounds better but I might be schizophrenic
I can't really notice a difference with feature retrieval rate
Also is crepe better than harvest?
pastel oak
Hi, I'm using RVC-GUI, I read that Crepe is the best one but I'm kind of confuse...
If youre new id recommend sticking with rmvpe instead of crepe, it is the best overall for any use case scenario. I personally only recommend crepe if you know what youre doing and an advanced model creator who knows their way with handling noise and all
tall rapids
I'm not training a model, I'm using someone elses models
rare gobletBOT
Ayo? @tall rapids level 1 !!!
tall rapids
these are just the settings I have in the program
I can't send a pic here
Suddenly I can lol
pastel oak
I would rather suggest you use Ilarias RVC for inferencing audio. What youre using is preeetty outdated
ilaria rvc
https://huggingface.co/spaces/TheStinger/Ilaria_RVC
guide if needed
https://docs.google.com/document/d/1YbXcLFPaGjhOdG5NFkK3QrucCEpHZBwFUxkeMO8aB18
pastel oak
Should technically be fine as long as you only infer vocals and not train your own model
1 Sec
tall rapids
yeah I didn't have problems with it yet
pastel oak
Mainline Local NVIDIA download link:
https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/RVC1006Nvidia.7z?download=true
Guide on how to use:
https://docs.aihub.wtf/rvc/local/mainline/
pastel oak
Daang youre an og
tall rapids
Yeah it was basic Pandas and Anaconda stuff
Facial detection
Racial detection
Emotion detection
It was way more lightweight
for example - attach a camera to a raspberry pi which is constantly detecting faces and counting how many of each emotion and each race you saw
sick fable
pastel oak
Yeah it was basic Pandas and Anaconda stuff
Yee things have come a long way and more user friendly than ever
Some user friendly stuff are hidden behind a paywall sadly but cant expect the best of the best for free, have to do some things manually
real field
which ia in google colab is best for training?
tall rapids
Some user friendly stuff are hidden behind a paywall sadly but cant expect the b...
yeah that's why I alwas go for locally hosted
rare gobletBOT
Ayo? @tall rapids level 2 !!!
violet heron
which ia in google colab is best for training?
RVC Disconnected
odd shale
which ia in google colab is best for training?
RVC Disconnected, Applio Colab or Hina's Mainline on Kaggle
(Kaggle got 30 hours of free use per week)
quasi iron
@odd shale Can you help me ?
rare gobletBOT
Ayo? @quasi iron level 1 !!!
odd shale
<@628603171735994379> Can you help me ?
Of course.
quasi iron
Of course.
I have a probleme with colab rvc
Thats working but when I use a model that not the same
I mean the voice doesn't like the model "I use a furina voice that not furina voice..."
odd shale
I have a probleme with colab rvc
Which colab are you using?
pastel oak
Popey is talking about realtime voice changer. Already helping in dms
odd shale
Popey is talking about realtime voice changer. Already helping in dms
Oh nice, go and help him then.
pastel oak
Guess my help is not good enough xd
odd shale
Guess my help is not good enough xd
Nah it's fine, you're already trying.
It's better than nothing.
pastel oak
UwU
mild quiver
can someone recommend me a voice model for horror narration
odd shale
can someone recommend me a voice model for horror narration
Take a look at the #1175430844685484042 channel.
worn river
guys, there is an error in the training at the epoch counter. im training in colab using titan pretrain. i tried a bunch of stuff:
- converted to mp3
- changed batch from 4 to 8 and back
- adjusted epochs from 450 to 300
... now what?
i.e i didnt train the voice model
night rune
-colab
worn river
ok thx
night rune
What happened to easyGUI?
worn river
What happened to easyGUI?
idk
worn river
guys, there is an error in the training at the epoch counter. im training in col...
applio did not train too
worldly oasis
did u run correctly the installation? <@274566299349155851> could u check ?
yes like everytime yesterday was okay
low shard
guys, there is an error in the training at the epoch counter. im training in col...
mp3 is bad, u should always use flac or wav, also u should specify the error and check if theres spaces or special characters in the dataset name
worn river
mp3 is bad, u should always use flac or wav, also u should specify the error and...
i always select wav, but i selected mp3 because maybe is some trouble with the dataset, the error happens in the training area in RVC V2 colab and only has names and numbers in my dataset name.
low shard
i always select wav, but i selected mp3 because maybe is some trouble with the d...
the applio colab? and what error
worn river
the applio colab? and what error
i didn't remind but i have closed the tab
i believe that is the dataset
low shard
no. https://colab.research.google.com/drive/1AoZMMoDGbQgnNcBUMgO7TO26Fg-jx483
where did u find that?
that looks like a portuguese version of rvcdisconnected
maybe try with https://colab.research.google.com/drive/1XIPCP9ken63S7M6b5ui1b36Cs17sP-NS and see if u get the same result
viscid moss
did u run correctly the installation? <@274566299349155851> could u check ?
hmmmm
weird
-colab
low shard
yes like everytime yesterday was okay
try to runtime -> disconnect and delete -> refresh the page and re run everything
wild vapor
-colab
willow lichen
Somehow when I submitted my "Goblin Caught on Tape" model to this server, I noti...
Thanks for reporting, we'll take a look
worldly oasis
Hi does anyone can gimme the output of one AI Cover i have the model i just need the final output but mine google collab isn't working it's about AI Cover
distant hamlet
its broken
let people fix it with one line of code lol.
#✨│ai-help message
worldly oasis
Please does anyone can just give me the output of one model that i have
pastel oak
Hi does anyone can gimme the output of one AI Cover i have the model i just need...
use ilaria rvc zero you can instantly use it without setting up anything
https://huggingface.co/spaces/TheStinger/Ilaria_RVC
worldly oasis
thanks i'll try
rare gobletBOT
Ayo? @worldly oasis level 4 !!!
pastel oak
worldly oasis
that 'looks to hard Google collab is more easy i cannot understand it i have already the full model i just need the output
pastel oak
You just insert the model url, download, go back to the other tab, load audio and youre done
you cant upload from google drive here thats the only thing ig
thin stump
that 'looks to hard Google collab is more easy i cannot understand it i have alr...
Pls don't fold the moment you see something new 
wintry palm
-colab
livid pilot
Ilaria basically broken like all the other colabs
Where do I find the guide to convert vocals locally
I installed the okada but thats real time
I dont want real time conversions
pastel oak
Where do I find the guide to convert vocals locally
Whats your GPU
livid pilot
Intel gpu
livid pilot
Whats your GPU
Intel
pastel oak
https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/RVC1006AMD_Intel.7z
After extracting, run go-web or go-web-dml if it exists
Follow this guide if needed
https://docs.aihub.wtf/rvc/local/mainline/
low shard
let people fix it with one line of code lol.
https://discord.com/channels/115926...
theres also another problem in the no ui btw, its not maintained overall
pastel oak
A few days ago someone said the go-web file is missing from the amd intel link actually so lmk if it is, so i can send it over
worn river
maybe try with https://colab.research.google.com/drive/1XIPCP9ken63S7M6b5ui1b36C...
ok i already woke up
low shard
Ilaria basically broken like all the other colabs
rvc?
livid pilot
https://huggingface.co/lj1995/VoiceConversionWebUI/resolve/main/RVC1006AMD_Intel...
Thanks
rare gobletBOT
Ayo? @livid pilot level 2 !!!
low shard
ok i already woke up
alr try
rare gobletBOT
Ayo? @worn river level 3 !!!
proven hill
Ilaria basically broken like all the other colabs
false
velvet folio
hey guys, im trying to make a voice model, but i cannot find the google drive folder im supposed to put it in
proven hill
not to sound an ass but why do you think it haves my name
heres the guide: https://docs.google.com/document/d/1YbXcLFPaGjhOdG5NFkK3QrucCEpHZBwFUxkeMO8aB18/edit
pastel oak
thank ilaria haha
not me:D
if theres anymore questions feel free to ask in here along the way
nono dont say sorry dww
proven hill
rvc disconnected and ilaria rvc (mainline in this case) works the same way, for the accent theres the (added) .index wich usually contains accents and much more, if youre looking to use this realtime / in a conversion i suggest you to higher up the index to make it better
odd shale
You can use Ilaria RVC Mainline and Applio Colab.
grizzled cove
Just a quick question, which one of these are the best? If it depends, which are used for what scenario?
proven hill
Just a quick question, which one of these are the best? If it depends, which are...
bs roformer viper 1297
grizzled cove
bs roformer viper 1297
Uhh
proven hill
Ilaria RVC -> Vocal Separator (UVR)
use the UVR standalone its better
grizzled cove
Where can I find it?
proven hill
my huggingface profile
rare gobletBOT
Ayo? @grizzled cove level 2 !!!
proven hill
Yep, thanks!
no problem!
quaint dock
hey guys, first attempt to train on rvc local, is it good like this?
I tried to process data and it said false
thin stump
I tried to process data and it said false
Don't train on Mangio. It's outdated https://docs.aihub.wtf/rvc/local/mainline/
grizzled cove
How do I get the higher tones right? It just goes to cackling or breaking down with my current settings. It also sometimes rapidly swaps from high to low.
(I'm covering MCR's "This Is How I Disappear" which has a lot of highs.)
Also what does this mean?
young belfry
is rtx 3050 6 gb good enough?
rare gobletBOT
Ayo? @young belfry level 1 !!!
quaint dock
Don't train on Mangio. It's outdated https://docs.aihub.wtf/rvc/local/mainline/
Alright, thanks!
thin stump
Alright, thanks!
also make sure you are using V2 only. It's not V2 if you don't see 32k
quaint dock
also make sure you are using V2 only. It's not V2 if you don't see 32k
alright, and 32k is the best setting or should I keep it on 40k?
rare gobletBOT
Ayo? @quaint dock level 10 !!!
quaint dock
aand can I also train locally with this?
and also like is it fully local like only I can see it?
thin stump
and also like is it fully local like only I can see it?
Your spectrogram shows your khz. 15 x 2 = 30k but still train on 32k. and yeah
adults cant hear 40k anyways
quaint dock
alright, thanks!
verbal egret
is there a updated version of RVC? Im using v.1.5.3.15
why does my voice sound so staticy when i play games and talk
rare gobletBOT
Ayo? @verbal egret level 1 !!!
violet heron
(I'm covering MCR's "This Is How I Disappear" which has a lot of highs.)
Based mcr
Make sure your using a audio with no reverb or anything
Turning down index rate might help
brazen skiff
Does anyone know if Weight works? The page does not load, a hots error appears
rare gobletBOT
Ayo? @brazen skiff level 1 !!!
violet heron
i want a voice changer
What’s your gpu
queen kraken
people say mangio is outdated, but isn't mainline also outdated since updates have been discontinued, except for applio?
vale osprey
-colab
violet heron
people say mangio is outdated, but isn't mainline also outdated since updates ha...
applio doesn’t have anything over mainline
Just because it’s getting updates doesn’t mean it’s better
queen kraken
okay so using mangio won't lead to any disadvantages, right?
violet heron
okay so using mangio won't lead to any disadvantages, right?
Mangio is bloated and would be slower then mainline
queen kraken
oh I see. what if I use mangio to extract pitch from mangio-crepe and then train with mainline?
analog obsidian
oh I see. what if I use mangio to extract pitch from mangio-crepe and then train...
if you want to use mangio-crepe use codename's fork, its based on the latest mainline update
(so is fast like mainline)
hot elbow
can you use more than one audio file for your data set when training
queen kraken
if you want to use mangio-crepe use codename's fork, its based on the latest mai...
okay thanks. ill check his fork
ripe hull
What should I do if there is no go-web.bat file after downloading and unpacking RVC?
livid pilot
false
Not false, it keeps giving errors when trying to export the vocals
And I wasnt the only one
violet heron
Not false, it keeps giving errors when trying to export the vocals
on the UVR tab?
If it’s exporting the audio that’s probably gradios fault and she can’t do much about it
ripe hull
thx
rare gobletBOT
Ayo? @ripe hull level 1 !!!
brittle wing
applio doesn’t have anything over mainline
Applio doesnt support Pre-Training?
Only mainline supports it locally?
violet heron
Applio doesnt support Pre-Training?
Only mainline supports it locally?
Pre-Training? Do you mean pretrains?
brittle wing
Pre-Training? Do you mean pretrains?
Correct
I don’t see an option while training within Applio
Locally*
brittle wing
you can replace the G and D files iirc
Within Applio’s main folder?
I drag and dropped those files inside the gui of Applio and nothing happened?
Told me to refresh and nothing showed up*
violet heron
Within Applio’s main folder?
I drag and dropped those files inside the gui of ...
replace where they were normally
violet heron
replace where they were normally
Should be the pretrain folder (?)
Should help you
livid pilot
If it’s exporting the audio that’s probably gradios fault and she can’t do much ...
I never blamed anyone, just saying no public converter is currently working
brittle wing
Should help you
Thank you
livid pilot
The ilaria hugging face is super inconsistent and slow
violet heron
The ilaria hugging face is super inconsistent and slow
what are you doing on it?
It should only take 30 seconds
brittle wing
Just because it’s getting updates doesn’t mean it’s better
Interesting, what made you think Applio despite of recent updates isn’t better than Mainline?
violet heron
It should only take 30 seconds
If your using it to separate vocals/instrumentals I highly recommend using the dedicated UVR space instead
livid pilot
It should only take 30 seconds
30 seconds is a long time
violet heron
Interesting, what made you think Applio despite of recent updates isn’t better t...
There is no point in it
They remove features then add them back claiming it’s new
It has forced RPC
brittle wing
True, you got a point there
violet heron
It has forced RPC
On multiple occasions “forgot” to credit people
violet heron
Rvc2 gradio used to do it in like 1 second
?!
I dunno what you were using but
No colab nor space ever got to that speed
livid pilot
Im not exaggerating if thats what ur implying
violet heron
No colab nor space ever got to that speed
Unless your files were very short
violet heron
Under 10-15 seconds
Then ilaira RVC should only take a few seconds then
livid pilot
Its taking 30 like u said
violet heron
Its taking 30 like u said
Nuhuh
livid pilot
So its a headache to use
livid pilot
Does ilaria on colab work?
violet heron
Does ilaria on colab work?
yes but the space is faster
livid pilot
Yea way too slow but also too many errors
violet heron
Yea way too slow but also too many errors
On colab or space?
livid pilot
Space
livid pilot
I need to get some local thing
violet heron
Literally just says error
Does the model name have any spaces?
livid pilot
Some convert some dont
violet heron
It’s inconsistent
Welcome to RVC
livid pilot
Whats the best local one
livid pilot
Welcome to RVC
Like i said the old ones were hella good
violet heron
Like i said the old ones were hella good
Then use the old ones
livid pilot
All the colabs are broken for the old ones
violet heron
All the colabs are broken for the old ones
Then don’t use it
livid pilot
Idk why they dont stick to those things tho and overcomplicate everything
violet heron
Idk why they dont stick to those things tho and overcomplicate everything
Ilaria RVC on HF is simpler then colabs!?
knotty moth
Whats the best local one
original RVC and Applio
livid pilot
Ilaria RVC on HF is simpler then colabs!?
Yes but slow
violet heron
Ilaria RVC on HF is simpler then colabs!?
And the reason we couldn’t stick to it is because google broke it
livid pilot
No its not + the waiting for gpu is extra time zzz
violet heron
No its not + the waiting for gpu is extra time zzz
Create a hf account
livid pilot
I have an account
violet heron
Creating a HuggingFace account is free
knotty moth
No its not + the waiting for gpu is extra time zzz
still faster than local inference in cpu mode but doable
livid pilot
still faster than local inference in cpu mode but doable
I dont mind waiting its just when it stops working its annoying
rare gobletBOT
Ayo? @livid pilot level 3 !!!
livid pilot
But thanks anyways yall
knotty moth
I dont mind waiting its just when it stops working its annoying
yeah you can do local one freely
livid pilot
yeah you can do local one freely
U said rvc?
I accidentally got the real time local vc
Instead of a regular converter
knotty moth
I accidentally got the real time local vc
run go-web.bat for mainline RVC
livid pilot
Mmvcserveriso
livid pilot
how tf did you accidentally download w-Okada instead of RVC
Cuz the applio one wasnt working so i tried finding another
violet heron
Cuz the applio one wasnt working so i tried finding another
Use the original
It’s still the best somehow
violet heron
where I find it
You need to open the install bat
If you want the original RVC https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/releases
violet heron
I just said if the model had any spaces in the name and it does
When you import a model make sure the model name has no spaces
violet heron
but was just testing
And it didn’t work because the model name had a space in it
livid pilot
mine doesnt but same thing happened before
violet heron
mine doesnt but same thing happened before
Ilaira recently restarted the space so it should unload the old models from the vRAM so it might work now
Just make sure when you import a model it has no spaces or special chatacyers
knotty moth
btw why u say make an account when i mentioned the connecting to gpu
for more GPU usage quota
violet heron
btw why u say make an account when i mentioned the connecting to gpu
More quota on the GPU and more priority connecting to one
brittle wing
replace where they were normally
Is this were I place PreTrainers in Applio Local?
livid pilot
If you want the original RVC https://github.com/RVC-Project/Retrieval-based-Voic...
its downloaded, what do i do now?
@knotty moth
torn mist
When generating audio with text, how could I give it more emotion and not sound like a robot?
knotty moth
its downloaded, what do i do now?
read the guide first https://docs.aihub.wtf/
livid pilot
read the guide first https://docs.aihub.wtf/
the go-web.bat is not here
there is only 1 bat which is for realtime conversions again
and I dont even know which guide to read
pastel oak
the go-web.bat is not here
did you download amd/intel prepackage version?
It is missing in that release for some reason. Heres the file
glass scaffold
Что такое epochs?
devout field
@pastel oak can u vc
devout field
env\python.exe: can't open file 'C:\Windows\system32\app.py': [Errno 2] No such file or directory
Press any key to continue . . .
help
low shard
Что такое epochs?
Я сейчас использую переводчик, эпохи — это единица измерения цикла обучения ИИ, нет подходящего размера, поэтому лучше использовать тензорную доску, как вы можете прочитать https://docs.aihub.wtf /rvc/resources/epochs--tensorboard/, также для русского чата используйте https://discord.com/channels/1159260121998827560/1159346439424573440
ruby pewter
can any one help me getting a older version off MMVCServerSIO cause the new one is really bad i cant find my graphics card option only gpu and cpu and thats really laggy and it was not like that before
rare gobletBOT
Ayo? @ruby pewter level 1 !!!
ruby pewter
pls does any one have a older version of the software
omg i got the good version yaa hope its better hihih
tall rapids
which one of these 3 should I use?
pastel oak
which one of these 3 should I use?
The model is the first one.
the D and G are for resuming training
patent pasture
May breathing and laughing being in the dataset affect negatively on how model sounds?
crude flame
May breathing and laughing being in the dataset affect negatively on how model s...
nope
patent pasture
nope
okay, how I understand the long pauses in the speech won't be extra too. Right?
crude flame
okay, how I understand the long pauses in the speech won't be extra too. Right?
wdym by "extra"
crude flame
kinda they won't affect on model badly
oh ok. Well it would be better to either truncate silence or use audio labels
patent pasture
I will know, thank you
tidal trail
Is this were I place PreTrainers in Applio Local?
go to download tab and download the pretrained you want to use in download custom pretrained
knotty moth
May breathing and laughing being in the dataset affect negatively on how model s...
not really, prob in reasonable amount, but quality, clarity, and variation matter
patent pasture
I will know, thank you
when I create someone's model, it often sounds not enough similarly to the original though I do everything possible to improve quality idk
hard
knotty moth
when I create someone's model, it often sounds not enough similarly to the origi...
voice similarity? prob not in the right pitch or the dataset is mixed with some other roleplayed voices
patent pasture
I got, thanks
devout field
does anyone have experience w applio
patent pasture
@nocturne mural , Applio was updated and so problem with arguments appeared in Applio_NoWebUI notebook. Can you fix it?
nocturne mural
ok
patent pasture
thanks
nocturne mural
<@989772388508000306> , Applio was updated and so problem with arguments appeare...
f5
patent pasture
you did it quite fast
brittle wing
any non local trainer with an ui
patent pasture
any non local trainer with an ui
I wouldn't recommend train using notebook with UI because (according to my experience) you can lose your progress in any time if it will disconnect
-colab
brittle wing
I wouldn't recommend train using notebook with UI because (according to my exper...
i searched up my older messages and found out i was using rvc disconnected
doesnt it disconnect only when i dont look at it once in a while?
proper shale
doesnt it disconnect only when i dont look at it once in a while?
it will actually disconnect after 4-6 hours of training, so yeah... be careful
proven hill
local option best option
crude flame
Just throwing this out there for any amd bois that it is 100% possible to train using amd on windows but it is a massive process to setup. If you want to find out how to set it up you can read this applio PR https://github.com/IAHispano/Applio/pull/513. You are going to need zluda and HIP SDK. This does work i am currently using it
brittle wing
**My output audio sounds robotic help? on rvc **
?
im only refering to the rvc mangio fork program
formal plume
noob question - i followed a recent youtube video to train my own 1 hr voice sample via tortoise tts...and get lots of extra artefacts in it...would using a pretrained model help with this? whats the difference between pretrained model and a voice model?
noob question - whats the difference between pretrained model and a voice model?...and how do i actually use a pretrained model on tortoise....i cant find a folder called pretrained v2 lol
polar meadow
I have a question could someone help me to create my first Ki song with my voice
odd shale
noob question - i followed a recent youtube video to train my own 1 hr voice sam...
A pretrained on RVC is the base you select for making a model.
If you try to train a model without a pretrain, it may won't sound good because you would be training from scratch.
odd shale
noob question - whats the difference between pretrained model and a voice model?...
Are you trying to use RVC?
odd shale
I have a question could someone help me to create my first Ki song with my voice
First of all, you need to make a voice model of yourself (your own voice)
polar meadow
First of all, you need to make a voice model of yourself (your own voice)
I've already taken 10 shots or so,
but the app or something doesn't work
odd shale
I've already taken 10 shots or so,
but the app or something doesn't work
Please, explain properly your issue.
!howtoask
patent trellis
polar meadow
Please, explain properly your issue.
So for me I want to do this with easyGui but I'm too stupid for the creation
odd shale
So for me I want to do this with easyGui but I'm too stupid for the creation
What do you want to do? Use an already made model for inference or make a new model?
Also, EasyGUI is outdated.
You can use Applio Colab instead.
low shard
So for me I want to do this with easyGui but I'm too stupid for the creation
the easygui google colab is broken #📰│dev-updates message as the creator isn't active since may
There's a fix but the colab is overall outdated, i'd suggest you to use the alternatives
polar meadow
the easygui google colab is broken https://discord.com/channels/1159260121998827...
ok
formal plume
Are you trying to use RVC?
yeah...basically i used this video (https://www.youtube.com/watch?v=7tpWH8_S8es) and a few of his other videos to download tortoise and train my own model from scratch....and the rvc voice model and index file i've been using are ones i downloaded from here.
fast forward to now...i see there are pretrained models. basically where do these pretrained models fit into the grand scheme of things? Is it in place of the rvc voice models i originally downloaded? if so, do i need an index file as well? .......or is it during the training process (where i took an hour long voice sample, split it, etc.).
rare gobletBOT
Ayo? @formal plume level 1 !!!
odd shale
yeah...basically i used this video (https://www.youtube.com/watch?v=7tpWH8_S8es)...
Nah don't watch that video.
Videos tend to be outdated.
Also, using community-made pretrains isn't really mandatory nor necessary unless you really wanna use any of them (Tho the majority tend to have issues OG pretrain doesn't have)
formal plume
Also, using community-made pretrains isn't really mandatory nor necessary unless...
understanding this reality very quickly lol
odd shale
understanding this reality very quickly lol
So, what do you wanna do?
Use a model to test it?
formal plume
Also, using community-made pretrains isn't really mandatory nor necessary unless...
i think you answered it with this....basically, i've been overwhelmed with "use this" "do that"...i just recently started this whole journey to help narrate some stories i have and was trying to get the best quality audio....even with the artefacts, it's wayyyyy better than what i used before lol
odd shale
i think you answered it with this....basically, i've been overwhelmed with "use ...
So, what you want to do is inference with an already made model?
formal plume
yes (sorry still learning the technical lingo)
odd shale
yes (sorry still learning the technical lingo)
Inference stands for converting an audio to the model's voice you're using.
low shard
Also, using community-made pretrains isn't really mandatory nor necessary unless...
I mean, pretrains can help for other languages as the original is trained only on english
odd shale
I mean, pretrains can help for other languages as the original is trained only o...
Of course.
brittle wing
I finished training with rvc disconnected, it showed me this and proceeded to not give me the pth files in the google drive folder. do i need to export it, is there some extra step i need to do to get the pth files?
odd shale
I finished training with rvc disconnected, it showed me this and proceeded to no...
Check your colab's file explorer
brittle wing
Check your colab's file explorer
how do i do that
patent pasture
I have trained model just, whats this?
brittle wing
After around 920 epoch, I got an error. What does it mean?
livid pilot
It is missing in that release for some reason. Heres the file
thanks a lot
analog python
hi, I am a newbie in Rvc, if I want to change an audio file to different voice by model download from voices-model. where should I start?
hollow geyser
trying to run applio locally, nothing happens when i run it. just opens a cmd window and does nothing else
i would send a picture but i seem to not have media perms so you'll just have to take my word for it
when i run it as an admin it says to run "run-install.bat" first, but that file doesn't exist
tight leaf
hi, give me a google collage notebook that makes ai covers + separates vocals from instrumentals, and is there an updated version of covergen?
trim yew
does the voice changer version affect my PC's performance? if so, can someone please help me to have a good version so my PC can handle it?
violet heron
does the voice changer version affect my PC's performance? if so, can someone pl...
What’s your GPU
trim yew
a MX-350
violet heron
a MX-350
Don’t think you would be able to run it well enough but if you want to try
Use this fork
wispy lodge
Aww man it didn't work
What's wrong tho?
trim yew
Even after trying it and making sure my mic is connected, I can't hear anything
rare gobletBOT
Ayo? @trim yew level 1 !!!
wispy lodge
Even after trying it and making sure my mic is connected, I can't hear anything
If you set monitor to speakers under audio settings, you don't hear yourself?
You can also screenshot settings
wispy lodge
Could you show the settings you use?
wispy lodge
It's not the version that you were linked
odd shale
hi, I am a newbie in Rvc, if I want to change an audio file to different voice b...
You can start using Applio Colab.
First of all execute the Install Applio cell and then Start Applio one and wait till a public url appears and click on it.
trim yew
And extract the one that ends with 001
do i put them both in the selected file and only extract the 001?
wispy lodge
put them in a separate folder
trim yew
it keeps stopping from downloading once it reach half-
wispy lodge
Then get it from here
https://drive.google.com/file/d/1UwsMJTzKOMPEmrF17iq7cjZpH1A-ppr2/view?usp=drive_link
patent pasture
I have trained model just, whats this?
I guess after an update the models made in Applio work only in Applio inference (I tested in the Ilaria Zero)
hollow geyser
how am i meant to install applio locally?? the docs say to run "run-install.bat", but both versions i've downloaded just don't have a file with this name
i've tried downloading an older version, tried showing hidden files, tried telling the command line to do it, but no luck for anything
queen kraken
how am i meant to install applio locally?? the docs say to run "run-install.bat"...
you can download zipped version https://huggingface.co/IAHispano/Applio/tree/main/Compiled
hollow geyser
that's what i downloaded, it doesn't have the file it says i need to run
queen kraken
I just downloaded, you can run run-applio.bat. no need to run-install.bat
hollow geyser
nothing happens when i do. it says to run run-install.bat
opening run-applio.bat doesn't do anything
if i run as an admin, it says this
queen kraken
weird mine works fine
tidal trail
if i run as an admin, it says this
do not run as admin
hollow geyser
running it not as an admin does literally nothing
tidal trail
running it not as an admin does literally nothing
hollow geyser
that is what i downloaded, yes
tidal trail
running it not as an admin does literally nothing
are you sure?
tidal trail
if i run as an admin, it says this
look
hollow geyser
yep, it just opens an empty cmd window
tidal trail
hollow geyser
?
hollow geyser
yeah, i know, i said even when i don't it doesn't do anything
hollow geyser
opening run-applio.bat doesn't do anything
look at this one
tidal trail
look at this one
wait some minutes
it will give you a link or open a windows in your browser
hollow geyser
i've left it open for multiple minutes, nothing happens
tidal trail
i've left it open for multiple minutes, nothing happens
weird it never happend to me
hollow geyser
ill try again and leave it for longer and see if it works
rare gobletBOT
Ayo? @hollow geyser level 3 !!!
queen kraken
where did you extract it? can you show the folder path?
hollow geyser
it. randomly started actually opening oh my god
it didn't even take a long time this time it just decided to work
tidal trail
nice
hollow geyser
i tried like 10 times i have no idea why it only worked now 😭
flint yoke
how do i run this thing bro
am on mac
people keep saying RVC realtime is good and allat but when i cant even use it (am on mac)
idk if am supposed to launch this thing called "MMVCServerSIO" but when i do after 3 seconds it says "python is damaged you should move it to the trash"
wispy lodge
idk if am supposed to launch this thing called "MMVCServerSIO" but when i do aft...
Did you download it from this link?
https://rentry.co/forkvoicechangerguide
flint yoke
well yeah
wispy lodge
flint yoke
alright thanks
flint yoke
Then follow this step
https://github.com/deiteris/voice-changer?tab=readme-ov-fi...
why is it this big???
o nvm (new to this sorry)
theres an error called "failed to execute client"
idk if that matters it might be
cyan sable
there shows no input after restarting my pc like i cant find my mic n everything
@wispy lodge
flint yoke
Then follow this step
https://github.com/deiteris/voice-changer?tab=readme-ov-fi...
sorry for being a bother, but ive run into 2 issues
rare gobletBOT
Ayo? @flint yoke level 2 !!!
flint yoke
for some reason, the voice is repeating itself
and it cuts off alot
like ALOT
it would be like he-l o- lo- lo-lo
cutting off with echo's
gusty crystal
that happens to me aswell
cyan sable
@gusty crystal u romanian?
flint yoke
when am i getting my answer 😔
wispy lodge
and it cuts off alot
Could you show your settings?
flint yoke
sure
cyan sable
i have this problem on the client but on the server it works, are there any fixes?
flint yoke
flint yoke
i have this problem on the client but on the server it works, are there any fixe...
no output and monitor on client?
yeah me too i had to use server lol
cyan sable
no output and monitor on client?
is there any fix
flint yoke
dont think so, okada has client tho
wispy lodge
i have this problem on the client but on the server it works, are there any fixe...
It doesn't show audio devices in client audio? Did you give microphone permission? There should be a microphone icon in address bar
cyan sable
It doesn't show audio devices in client audio? Did you give microphone permissio...
yeah
thx
flint yoke
so?@wispy lodge
wispy lodge
so?<@277821614345945089>
How much perf does it show when you speak?
flint yoke
220-275
wispy lodge
Well, echo is usually caused by microphone catching the sound from speakers. So either enabling noise suppression or turning volume down should help
But regarding cutting off, it shouldn't be if perf is fine...
Is it Safari? Could you try opening the link in Chrome and try client audio mode there?
flint yoke
Well, echo is usually caused by microphone catching the sound from speakers. So ...
you know, i would if i wasnt locked out of the client thing
(like the dude above)
but yeah it opened automatically on safari
chrome didnt help 💀
wispy lodge
Don't see audio devices in client mode in Chrome?
worldly totem
help me idk but my voice changer not working yet and settings cant help me
wispy lodge
It should prompt for mic permission
worldly totem
It should prompt for mic permission
help me please
flint yoke
Don't see audio devices in client mode in Chrome?
oh, i can use client now
whats better sup 1 or sup 2
wispy lodge
whats better sup 1 or sup 2
Sup2
worldly totem
my voice changer cant see my gpu, i have intel iris xe
flint yoke
YOOO THERES NO CUT OFFS NOW
tysm
even though it doesnt sound markiplier at all atleast it sounds decent
theres some cut offs at the end of my speech but idk if its universal
wispy lodge
my voice changer cant see my gpu, i have intel iris xe
Iris Xe wont work well unfortunately. You can try fork, but delay will be like 2-3 seconds
https://rentry.co/forkvoicechangerguide
worldly totem
😩
wispy lodge
theres some cut offs at the end of my speech but idk if its universal
iirc that's how sup2 works. You can try reduce In. Sens. to default -90 so it catches quiet audio better since noise is removed
And you can try increasing extra a bit to improve speech quality
worldly totem
i have good cpu, i5 1155g7, or it nothing change?
rare gobletBOT
Ayo? @worldly totem level 1 !!!
worldly totem
brittle wing
Anybody know how to fix
'No model for model_common loaded. Please confirm the model uploade
wispy lodge
i have good cpu, i5 1155g7, or it nothing change?
Haven't tried with it. Performance with CPUs is usually not great, but I had like 1.4s in total with ryzen 7 5800h. And that's ~80% utilization
worldly totem
Haven't tried with it. Performance with CPUs is usually not great, but I had lik...
I don't care about the delay as long as it works at all, I have new components that should work fine, it seems to me that I downloaded the wrong version, is this possible?
flint yoke
why can i hear echo's even with sup2 enabled?
fixed echo, found a new problem
now it suddenly cuts most of the time, please help
rare gobletBOT
Ayo? @flint yoke level 3 !!!
flint yoke
ignore my angelic voice
granite hare
no matter the model and tune my voice dont change
violet heron
no matter the model and tune my voice dont change
Is passthru red?
granite hare
Is passthru red?
yes
granite hare
Make it green
oh alr ty
violet heron
for some reason the voice changer wont work
!howtoask
rare gobletBOT
Ayo? @quasi night level 1 !!!
violet heron
hey guys, whats the minimum good amount of epochs
There’s no set on “good”
Try 200 and see if it works
quasi night
What do you need help with
ive set everything up (including the audio cables) and when i test the voice changer no audio comes out
violet heron
ive set everything up (including the audio cables) and when i test the voice cha...
Did you click the green start button
quasi night
Did you click the green start button
yea
quaint dock
There’s no set on “good”
alright, thanks
quasi night
(not in the screenshot)
quasi night
What’s your GPU
violet heron
Send a screenshot of the full voice changer
Including input/output/monitor
Everything
quasi night
ok
lemme just check rq if i need anything to cesor out for privacy reasons
here ya go
quasi night
Set monitor to your headphones/speakers
alright
kind bluff
here ya go
alguien sabe como descargar la ultima version?
quasi night
Set monitor to your headphones/speakers
tried that
still didnt work : /
and when i tried passthru
i just heard my regular voice
violet heron
i just heard my regular voice
That’s because it’s passthru
quasi night
That’s because it’s passthru
oh
quasi night
That’s because it’s passthru
any other fixes you know of?
violet heron
any other fixes you know of?
Try increasing out
violet heron
here ya go
Wait make your input your microphone
quasi night
Wait make your input your microphone
ok
quasi night
Wait make your input your microphone
tried both
rare gobletBOT
Ayo? @quasi night level 2 !!!
quasi night
still not working :/
dire bridge
Good evening guys, I have an RX 7700XT and I downloaded a voice program that is not compatible with my card unfortunately it only recognizes Vega 8 and RTX 3060, does anyone have a tip on a program I can use? please send the link now if possible
gray tinsel
where do i get RVC v2? i'm only seeing v1 stuff, do i need pitch tracking? whatever that is, we have a model info tracker?
granite hare
anyone know where to find really good ai voices like paid ones etc
knotty moth
I wouldn't recommend train using notebook with UI because (according to my exper...
nah, you won't lose the progress in hina's kaggle notebook, you prob forgot this persistence option
brittle wing
Hello
is there where to ask for help?
i have this problem
where it says pipeline not intialized
flint yoke
why is my voice normal on the website
but when i play a game it gets super laggy
like when i use realtime rvc on the website: "test test"
when i play something: "t-es... st s"
analog obsidian
but when i play a game it gets super laggy
windows gives more gpu resources to the game, which makes the voice changer lag, this is normal windows behaviour
you can fix the first issue by increasing chunk size (this also increases delay)
try 128 chunk size and above when playing games
flint yoke
windows gives more gpu resources to the game, which makes the voice changer lag,...
am using mac
analog obsidian
am using mac
similar issue, mac gpu (or whatever mac uses lol) is too weak to keep up with the voice changer + the game
increasing chunk size is the only way to fix that
flint yoke
increasing chunk size is the only way to fix that
how do i do that
oh you mean the chunk thing?
analog obsidian
how do i do that
u using the modded version right?
flint yoke
like the slider
flint yoke
u using the modded version right?
uh
analog obsidian
yea just increase the slider
flint yoke
am just using realtime rvc
analog obsidian
can you
analog obsidian
wheres that
show me a screenshot of whatever are u using lol
here in this channel
flint yoke
flint yoke
uh
last time i increased it my ears bleed
like it was so damn loud
literally sounds like null from baldi
analog obsidian
wat is not supposed to do that
flint yoke
ok tysm
theres no more chops or whatever its called
delay got increased but eh
small price to pay
brittle wing
Does anyone know how to fix where the system says pipeline not initialized?
it seems like the application voice wont run until it initialized?
I followed a tutorial on everything but the the voice just wont come on, either the demos or any models i downloaded
Help on this?
gray tinsel
why is my voice normal on the website
which site can you test the voice?
open stag
- colab
main tinsel
deiteris voice changer issue.
It seems like the modulation of deiteris' voice changer became strange at some point. What I checked was version b2174 later. When you play the latest version, even though it is the same model, the sound is a bit more bizarre and you hear more mechanical sounds.
analog obsidian
what issue u have
rvc or wokada
send more messages until u level up
is like 3 or 5 messages iirc
1
rare gobletBOT
Ayo? @brittle wing level 1 !!!
analog obsidian
nice