Description
DRIVE - HUGGINGFACE A much higher quality HandUnit Dataset, trained on RVMPE at 350 Epochs. Uses nearly all lines from Sister Location, Help Wanted and Help Wanteds DLC: Curse Of Fredbear Dataset roughly around 20-25 Minutes long. Credit to for helping with de-verbing / de-echoing the dataset, since all his voice lines have echo since theyre played through speakers and such in the games. If used, please just credit my discord :))
Comments
SPEAKING EXAMPLE -
SINGING EXAMPLE 1 -
SINGING EXAMPLE 2 -
pretty sure the voice actor said he didnt want his voice cloned
great model though
AI Hub models do not fall under DMCA, so everyone can have their voices cloned here.
yeah we can TECHNICALLY legally make models of anyone but like
dude doesn't want his voice cloned
More of a grey area. And it's gonna be an interesting thing in the future concerning laws involving all this. Like what if people say their AI model training data just came from someone who really good at imitating artist XY
my point is i personally respect people not wanting their voice cloned, if they don't want models made of em then we shouldn't make models of em
simple as that
Probably AI models will be created to determine if a voice actually belongs to someone or not, but then.. the models created here rarely have 1:1 accuracy, so the argument could hold up easily
Sure understood, wasn't trying to debate anything lol. My mind just started wondering about the future a little
They should hire us to check for inaccuracies to make sure it's not misused.
Like Sam Altman we offer the solution to the problem we created
Now everyone on this server has a job.
"Connection Terminated. I'm sorry to interrupt you Elizabeth, if you still even remember that name"
The rare good fnaf model
nice
what about the teen voice
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
Weekly Metrics
Users also tried
More to explore
Loading more