He adds formant on his voice in the songs with this voice, if you are making a song u should be fine with not adding formant becasue its trained off of vox with the effect however i would recomend a small value like 95
#Playboi Carti (2022 Era) [RMVPE] [RVCV2] [550 EPOCHS]
1 messages · Page 1 of 1 (latest)
Model Ready ✅
This model has been synced with Weights and is ready to use for free!
Bro I swear if someone pays you for these models, you guys gotta stop buyin it 😭
bro u have no clue what the voices are
Well I do but the quality isnt worth paying for
Provided to YouTube by IIP-DDS
2 Out · Iᐸ3PLAYBOICARTI
mixtape
℗ jordan carter
Released on: 2023-11-11
Composer, Lyricist: Jordan Carter
Auto-generated by YouTube.
skip to 19 sec
its the exact same
there are only 4 songs with good enough data to extract and for all of the voices i have on here i have the best models for them
they arent gonna be the best models ever because all of the vocals are extracted from songs with heavy effects but they are the best it can get and i make them beacause people avoid training theses voices to avoid ppl like u
People like me?
I dont have an issue with any one else in this server and I never said the model had to be good
but charging people 10-15 bucks for the quality of these is a low level scam
also I might have had a stroke trying to read your paragraph. It makes zero sense.
Also... I think only model masters are allowed to take model comms. (for obvious reasons)
In fact @reef valve you can't take comms if you don't have model master role.
okay I deleted that from the thing that's my fault but not my point
No bud, i'm not asking anything about your Carti models.
Oh of course, you can't put messages like that on your model posts if you don't have the model master role.
Sorry for repeating this. 🐢
ur good I just didn't know so thank you for letting me know
I'm all for constructive criticism but when u sob the post and then just say genuinely mean things I'm gonna defend myself like if u think it's so bad be nice and tell people how they can do better
I'm tryna get ready for school bc I have exams mb
so what is bad about this model so I can improve
for ppl who are thinking abt using the model here's killers ran thru it
definitely not a talking model but idk who is gonna talk with killers voice carti😭
The inside of his mouth glowed like Freddy Fazbear's eyes after the power goes out
maybe he is Freddy
this is my first applio model and lowk it turned out better than my others (relative to the dataset ofc) so I can't wait to train my next model from studio stems instead of low quality leaks to really see how good applio is and my data cleaning proccess. I feel like a lot of people think I'm not good at training models and that's understandable when u don't see the data that I have to use for the models I've posted, my baby voice model comes from Instagram and yt drops from carti which as yall prob know are stored in ogg format and on top of that their is only like 5 min of those songs and they are all super noisy, this model uses 5 min of data bc again there's not many leaks with this voice and they all have heavy effects on the voice. honestly I think I did p good tho with these in comparison to other models that use the voices that I've trained so far
I got no excuse for the deep voice model tho bc I didn't deess the data😭
but ya next model gon show my true potential on some underdog type shit bc it's gonna be off clean hq studio accapella
whole lotta yap fr but good insight on my line of thinking
I use collab bc it's easy :p
me too but the applio colab is 10000x better than rvc disconnected bc of the discriminator (and high end mirroring according to razer) but the models come out so much better
plus ur drive storage doesn't get filled up and u can export models while training
Sorry I didn't mean to sound mean last night. It was like 2 in the morning and my brain was not braining. The model is good now that im listening to it when im awake
Mb