🔒 Voice Cloning

[chrome marlin]

Let's say I have some mp3 files and some transcriptions of those files - how do I then train a model so that I can have a voice cloner? Can you please help me? I have looked into it a bit I know one can use e.g. Mozilla tts etc, but I just dont know really actually how to do it?

[mystic pawn]

@chrome marlin

Python help channel opened

Remember to:

• Ask your Python question, not if you can ask or if there's an expert who can help.
• Show a code sample as text (rather than a screenshot) and the error message, if you've got one.
• Explain what you expect to happen and what actually happens.

:warning: Do not pip install anything that isn't related to your question, especially if asked to over DMs.

Closes after a period of inactivity, or when you send !close.

[graceful valley]

Theres a simple way that runs a model locally

I know hugging face has a bunch of models that are easy to use

[wispy bluff]

this sounds ominous as voice cloning with ai/ml can be used for many malicious and illegal things
i think this might be leaning towards rule 5 here

[graceful valley]

Anything that involves code could be used for malicious purposes. But most voice clonings are mostly obvious that its a voice clone.

[chrome marlin]

...

there is nothing illegal about what im trying to do I assure you

[graceful valley]

Most voice clonings in python can only take one audio file so you’d need to combine the audio files

[chrome marlin]

and the ppl who are gonna wanna do illegal stuff, dont worry they wont have to ask on a discord server

xd

[chrome marlin]

gotcha

thank you

oooh okay

that's totally fine i can combine them and transcribe them together

i want to run it locally indeed and then be able to generate an audio version of a text file, in that voice

how do i do this where do i start with a simple way to do all this?

Should I look into hugigng face?

i'm on a mac system btw if that makes any diff

[graceful valley]

I used hugging face

But some times things dont 100% work so make sure to check the comments of the model

I already have some code that can clone voices from an audio file

But im unfortunately not home atm

[chrome marlin]

daym

if I may ask, would you be willing to share the code with me?

[graceful valley]

Yeah ofc

[chrome marlin]

that is very helpful.. thank you!

I'm in no rush, and much obliged

[wispy bluff]

which can be used for impersonation and can turn very dark very fast

[graceful valley]

But most voice clonings dont have good enough quality to trick impersonate

[mystic pawn]

Python help channel closed

This help channel has been closed and it's no longer possible to send messages here. If your question wasn't answered, feel free to create a new post in #1035199133436354600. To maximize your chances of getting a response, check out this guide on asking good questions.