Description
Dataset prepared and trained by me Model training based 80% from Stripped album Epochs < Epochs <
Comments
Sample: (500 Epochs)
please train it to 1k epochs so the output is cleaner
this model is so good
I will once i am done with Avril's <:calmcryandsmokin:1093342214765150399>
you are doing avril too π³
niceee
Christina Aguilera (RVC) 800 Epochs
Could someone please try the 800 epochs β apparently many said that this sounds nothing like Xtina and β it came out sounding like young Britney Spears instead ππ
it does sound like her, but many people knows the loud xtina better
Then i guess i chose the wrong input (Britney's dry acapella) as it doesn't showcase Xtina's full range enough ππ»ββοΈππ»ββοΈππ»ββοΈππ»ββοΈ
try whitney houston
while you're at it, please make sure you don't transpose incorrectly
because the sample is +3 higher than it needs to be
set the feature index to 0,90 instead of the standard 0.76 when inferencing
and you will see it will sound more like her
oof thanks!! should i continue to train it to 1k?
nah
Sample: (800 Epochs) generated by Princess Aurora
Hi, your xtina AI is really very good!!! but the huggingface links are not working, neither the 800 nor the 500 one, it gives an error due to the index ''Error index -1 is out of bounds for axis 0 with size 0'' in aicovergen
Add a comment
Samples
1. Singing
Male
English
2. Singing
Female
English
3. Singing (Dry)
Female
English
4. Singing (High)
Female
English
5. Singing 2
Male
English
6. Singing (Dry)
Male
English
7. Singing (Dry, High)
Male
English
Pitch
Users also tried
More to explore
Loading more