- Joined
- Dec 28, 2014
The Palantir connection is much more interesting.
Elevenlabs.io - AI Generate anyone's voice with a minute of audio saying how much they hate trannies
- Thread starter Colon capital V
- Start date
-
🐕 I am attempting to get the site runnning as fast as possible. If you are experiencing slow page load times, please report it.
- Joined
- Jan 4, 2019
I have been playing around with this all week since tuesday its so addicting
biggacracka
kiwifarms.net
- Joined
- Dec 14, 2022
Inspired by the deepfake of Linus Sebastian with Terry A. Davis voice, I tried to transcribe the quote to text and feed it to the AI with Linus samples
rareblacklobster
kiwifarms.net
- Joined
- Feb 7, 2021
Does anyone have any Mario voice samples that work? The ones I've tried always give me some normal, faggy sounding voice.
- Joined
- Oct 7, 2020
Have any of those Dumbledore-posting scripts been read out with Dumbledore's voice? That would really make some trannies REEE.
- Joined
- Mar 4, 2019
- Joined
- Apr 17, 2022
Does anyone have a download so they cant retroactively impose censorship on this thing?
- Joined
- Aug 7, 2017
I have a question - it's obvious that videos where Prince Harry tells "Britain should cleanse itself of shitskins" are gonna be deleted, but what if I upload a video where this AI reads in some person's voice what they actually tweeted, posted, etc, if it doesn't have any naughty words or implications? Can it be taken down for some gay shit like "defamation" or "character assassination" or it can be considered transformative work? I mean, it's YT, so it can be taken down for whatever reason they will pull out of their asses, but still.
- Joined
- Mar 30, 2021
Palantir in the UK. Covering the time Dominic Cummings, was ruling the roost in 10 Downing St. (Though Carrie Symonds got him fired, lol) which links the whole thing back to Cambridge Analytica. Interesting indeed.
- Joined
- Dec 28, 2014
It almost makes me wonder if it actually is a glowie op to start hysteria about AI because the fact this would inevitably be used to get Emma Watson to rant about niggers or whatever was obvious. The glowies want AI bots for their own disinformation use, not for the plebs.
- Joined
- Mar 30, 2021
It's an excuse for more censorship of the internet, and open source apps, perhaps. Ties in with incredibly overreaching EU nonsense.
We'll see soon enough, I suspect.
Between Josh's predictions and the rest of the speculation on the farms, it's starting to feel a bit Cassandra like around here.
No wonder everyone wants the whole place gone.
Crowd sourced information gathering and expertise could prove dangerous.
- Joined
- Feb 24, 2019
It's interesting that it seems the more you know a voice the less it works. Video game stuff is the best, I know the voices but haven't listened to long segments of it and infrequently for years.
However, those I've listened to long form content for years don't work at all. Metokur is barely passable as a attempt at him, Null doesn't work at all and I'd never make the connection absent of context.
However, those I've listened to long form content for years don't work at all. Metokur is barely passable as a attempt at him, Null doesn't work at all and I'd never make the connection absent of context.
- Joined
- Oct 28, 2021
Biggie raps Moonman
It doesn’t sound like biggie to much (especially at the end) even though I used the acapella to hypnotize
biggacracka
kiwifarms.net
- Joined
- Dec 14, 2022
It might also be because video game characters have mostly the same "character voice" even through different emotions and are made to be consistent, recognizable, while stuff like let's say podcasters can range from different mood and emotions and can be sick or other so their voice can vary a lot more, and there's no double take for live content.
I think if you were to take voice clip from and for movie characters instead of the actor, you would get results closer to the video game stuff.
- Joined
- Feb 13, 2018
- Joined
- Dec 3, 2020
KingLiberty76
kiwifarms.net
- Joined
- Apr 26, 2022
I don't remember this video by Seth
- Joined
- Jan 31, 2021
The problem with Null is that it can't really capture his nasally delivery. It ends up pitching him up and making him sound like a fucking child (or like he REALLY likes anime). I haven't heard any Jim deepfakery so I can't speak to that. Got any samples?
- Joined
- Feb 24, 2019
It's in the OP.
Metokur I think does sound like him with no inflection, however Null's is the worst in this entire thread, even Americanized Dankula sounded more akin to it's source than all the Null's.
Maybe that'll be a marketable trait in the future, having a vocal style (or even artistic one) that's hard for AI to recreate.
- Joined
- Dec 31, 2022
If you're technically oriented, check this out:
AI Voice Cloning
This repo/rentry aims to serve as both a foolproof guide for setting up AI voice cloning tools for legitimate, local use on Windows/Linux, as well as a stepping stone for anons that genuinely want to play around with TorToiSe. Similar to my own findings for Stable Diffusion image generation, this...
rentry.org
I've trained it to talk like Null with approximately 30 seconds of the latest MATI, it's pretty good, but I have an AMD GPU, so performance is abysmal sometimes.
God bless FOSS developers.