I can't find the index file

[hearty slate]

GPU : Nvidia RTX 2060 Super
16 Go ram
WIndows 10

I've followed all steps until generating index file, but impossible to find it.

Tutorial used : https://docs.aihub.gg/rvc/local/applio/#extra

Can someone knows how to fix that ?

[strong grove]

Always make sure both Applio RVC and your NVIDIA GeForce Game Ready driver are up-to-date.

[hearty slate]

Hi ! I just saw that I was using the Nvidia software instead of AMD...

I've planned to continue the battle of installing but after reading "These guides are for AMD GPU users on Windows using Zluda to enable CUDA compatibility" my bravery was vanished 😂 I give up.

It's too complicated for the newbie that I am, I think going on ElevenLabs and pay my 5$

[strong grove]

You have stated your PC GPU is NVIDIA GeForce RTX 2060 SUPER. Where did you say you have the AMD Radeon RX GPU? If you give more time to get Applio RVC to work with your NVIDIA GPU, it would be great, just skip that specific "AMD GPU" part on Applio doc if you have NVIDIA GPU and not AMD one; otherwise it's your loss to give up at first glance.

[hearty slate]

You're really kind and I thank you for that. The fact, is that I've spent many hours yesterday with a big headache in bonus...

I've followed all the guide for Nvidia, but my path is always blocked by this sentence in the console that you can see on the picture that I've send above, impossible for me to create the Index...

Where I can see if the Applio driver is up-to-date please ?

[strong grove]

https://huggingface.co/IAHispano/Applio/blob/main/Compiled/Windows/ApplioV3.6.0.zip

[hearty slate]

Yes, this is the one I used

Look, this is a part I don't understand.
Context : I want to create a voice model by sending samples, in a way to do TTS.

Why they are asking me for .PTH and .INDEX at the beginning ? I don't have those

@strong grove

[strong grove]

RVC is not a TTS-type software; it's more of speech-to-speech (STS). By saying "to create a voice model" it means you want to "train" a complete RVC voice model. This part of the guide states to upload an already-existing RVC voice model, which typically found in #1175430844685484042. It's not a training one.

[hearty slate]

This is the right guide for my needs ?
I want to make some sentences said by the voice character I want, let's imagine : Mario saying "Hello, welcome here"

[strong grove]

Are you sure RVC itself can do TTS? Not always. Though in Applio RVC, there's a feature named TTS. https://docs.aihub.gg/rvc/local/applio/#tts

But first of all, you would have to upload two files, pth and index into Applio, these are voice model files.

[hearty slate]

How do you do that ? It's the part that I don't understand

[strong grove]

In my first screenshot, these files are of "RVC voice model"; they both typically come within a zip file. You use 7-Zip or WinRAR to extract these files into "Applio\logs". Go on Applio RVC GUI, click on "Refresh" button to load voice models, and there you'll see some uploaded voice models appear.

[hearty slate]

Yep, got it. But, the voice model that I want doesn't exist, I need to create it, that's the part that I don't understand, how ?

[strong grove]

Now that I have to tell you every step because you don't know what to do with Applio, where I somehow know better but I'm just complex at explaining.

[hearty slate]

The only part that I don't understand is how to create my own RVC model

[strong grove]

This section on Applio is "Training". As what I said earlier, the term "training" simply means to create your own voice model, as what you understood.

[hearty slate]

ooooh

[strong grove]

How do you know to prepare your "dataset" that contains audio file(s) to "train" in Applio RVC? That part where I question about your statement if you mean something else.

[hearty slate]

For training, I've prepared a audio of 2min of my character's voice

[strong grove]

To train your own RVC voice model, check this https://docs.aihub.gg/rvc/local/applio/#training, this https://docs.aihub.gg/rvc/resources/dataset-isolation/ and this https://docs.aihub.gg/rvc/resources/training/.

2 minutes of dataset audio is quite low, although it's possible to train a voice model using dataset audio of lower than 10 minutes with some settings, the sound at the result (making AI cover) would also depend on your dataset too.

[hearty slate]

Unfortunately, that's all I have for my character 😕

[hearty slate]

At first, even TensorBoard is impossible to install. I've downloaded it, put it on RVC's folder, and I've run it like the guide said, and it does'nt work

[strong grove]

If Tensorboard isn't built within Applio, do you have any idea what's this bat file even for?

[hearty slate]

Yes I've read the guide abotu TensorBoard and browser

[strong grove]

If there's a terminal after you double click on the Tensorboard batch file, copy the url link and paste it in the same browser where Applio RVC is also displaying at the time.

[hearty slate]

Yes, but what i'm saying is about this part