This is a retrain of my DanTDM 600 epoch model
Trained & Tested on Applio(Kaggle)
Talking model cannot handle singing voices
Used RX 10
De Crackle
De Plosive
De Ess
Voice De Noise
Normalize -3.00
Dataset: 14 Minutes 56 Seconds
https://huggingface.co/notiscBalbalEko/myrvcmodel/resolve/main/DanTDMRetrain_1300e_11700s.zip?download=true
#DanTDM(Retrain - Cleaned)(RVC V2)(Original)(RMVPE)(32k)(1300 Epochs - 11,700 Steps)
48 messages · Page 1 of 1 (latest)
Rate the model
To rate this model, either use /rate or click the buttons below.
U used batch size 16? Epochs are too high for that dataset i think
it was on 12
and ofc its gonna be high cuz i trained this on 1300 epochs
sounds kinda muffled, but good job. you shouldn't use de plosive (ruins your audio), neither voice de noise (not necessary, mel denoise/roformer can do that better)
Idk half the things used to clean the audio, I only know dereverb/echo, dresser, and denoise
dresser?
you mean de-esser?
Oops
Ayo? @long crane level 32 !!! 
Yes
t de-ess right?
What
dont tell me your using that instant dataset maker
Absolutely not that thing is yucky
Oh ok then, what de-esser you use?
huh
Ask razer about t-de ess
it's a good and free de-ess
How do I use it? Could you send
Idk the settings, but ask Razer as i said before
@dark pecan sorry for ping but can u help him about t-de ess plz
Here is the website where you can get it: https://techivation.com/t-de-esser/
Here are some good settings for it
it did get updated so the ui looks new but the settings are the same
sure
How do I use this?
I'm so confused rn 😅
fl studio
Lol what is that noise
glep
Model Ready ✅
This model has been synced with Weights and is ready to use for free!
