Can someone help me with this two things if its possible? :(

[lethal turtle]

Hi! So, i had never used AI and i have 0 experience with programming stuff... but i was talking with a friend and he kinda suggested me to ask for help here. Basically my situation is that i want to make 2 things:

1.-A Voice model of my own voice: This is bc im not an english speaker, i speak spanish. I want to do stuff in English but i always have problems pronouncing stuff, so my friend told me i could try to get a VM of myself so i can do the things i want to do without having the pronunciation issue there anymore. I would like it to be kinda a translation but is okey if its something difficult to make.

2.-Text to voice in a Videogame: Funny, um...basically im part of this proyect but again, language issues will happen and i was asked to find a solution since probably most of the ppl were going to stream their things and i dont want to cause issues bc of them not understanding me (yeah, kinda sucks) and again, my friend told me i could ask for help to make one so i could write something and in the game or in a discord VC and the other person could hear the text i wrote down.

[true quest]

The accent thing only works if you have someone who is a native english speaker record the thing you want to say, and then convert the audio he sent you to use it with your own voice just so youre aware

Pretty simple to create a model, has nothing to do with coding either:
https://docs.aihub.gg/essentials/how-to-make-voice-models/

Dataset = clean vocals of you talking, 15 minutes is enough, no background noise and stuff, volume shouldnt vary too much - it shows how to clean your audio aswell

RVC - if you have an RTX Nvidia gpu, use local option with Applio, else on the cloud which will be a bit more challenging

Go to the tensorboard step in extras aswell and get that to determine when the model is finished

Ehh I think wok's tts client works as realtime, but theres no guide or documentation of it done yet aside from what wok did
https://github.com/w-okada/ttsclient

For the tts client youll need a different kind of voice model architecture which is gpt sovits, explanation on how to train that someone else will have to help

[lethal turtle]

Whats an RTX Nvidia gpu and an Applio?

[lethal turtle]

ooh i see i see

[true quest]

If you dont know what the most crucial component of a computer is im worried

Applio is a software to create voice models, make ai covers in post production

[lethal turtle]

aaaaaaaaaaaaaaaaaaaah that thingi

yeye, i know about them, i just kinda remembered them with other name

[lethal turtle]

i see

[true quest]

Well thats the newest one thatll be releasing, you dont necessarily need to invest 3k+ for just a gpu

[lethal turtle]

jsjsjsj probably, i mean i cant buy a new one so i hope my pc can make it without exploding

[empty elbow]

you can find an affordable RTX 3060 in some used market

[lethal turtle]

nono, is more an authorization to buy one, and since my pc is still working without problems, it will be impossible to get a new one

[true quest]

Destroy it

[lethal turtle]

[true quest]

If its not an old pc you can just replace the gpu

[empty elbow]

[true quest]

Simple as taking out and put it back in

Depends on your other pc parts

[lethal turtle]

is a laptop?