Elevenlabs.io - AI Generate anyone's voice with a minute of audio saying how much they hate trannies

  • 🐕 I am attempting to get the site runnning as fast as possible. If you are experiencing slow page load times, please report it.
The main idea I've always had was Raziel from Legacy of Kain, reciting the lyrics to Material Girl. Mostly due to my child autism brain always associating the two when I played it as a kid because of all the mentions of the Material Realm. Too much of a poorfag to spend money on a meme though
 
The main idea I've always had was Raziel from Legacy of Kain, reciting the lyrics to Material Girl. Mostly due to my child autism brain always associating the two when I played it as a kid because of all the mentions of the Material Realm. Too much of a poorfag to spend money on a meme though
Well there's Tortoise if you want something that somewhat works.
 
True, though I gave it a look for a few minutes before remembering I'm a tech-illiterate retard, so maybe not. I dunno, I might make another attempt in the future, we'll see.
Tortoise sucks, you wait five times as long for results that sound nowhere near passable. With the 11.ai on the market everything else really just doesn't compare anymore.
 
  • Like
Reactions: Puff
I decided to check on it to train it on Daniel MLG's voice, but it's completely pay-to-use now. That's a shame, but it's fair, I guess.
Now I'll have to get a new PC with an Nvidia GPU and install Tortoise TTS.
 
Not sure which model this is, but someone used AI to make Taylor Swift sing an animu opening. The fidelity is amazing.

 
Anyone got the fast track to Tortoise? I don't need voice cloning, I just need something that can read my novel in an old woman's voice without spending 400 dollars.
 
  • Feels
Reactions: Cowboy Kim
I'm seeing YouTube videos of AI generated voices with US Presidents. How do people do that? I think it's cloning.

I want to try that with Joan Crawford or Ingrid Bergman.
 
It's usually sampling a lot of things the subject says and feeding that into a learning model, so if you have an archive of Crawford or Bergman speech it might be possible to create a model based on them.

There's a lot of open-source projects on GitHub, but beyond that I don't know which ones are good.
 
I'm seeing YouTube videos of AI generated voices with US Presidents. How do people do that? I think it's cloning.

I want to try that with Joan Crawford or Ingrid Bergman.
There's a thread on AI singing here that people seem to use a Google Collab link for, with models from Huggingface or whatever Discord MEGA link they have to a character's voice model. Shit kept breaking on my end and couldn't get it to work for some reason but maybe you'll have better luck if you wanna mess around with it. https://kiwifarmsaaf4t2h7gc3dfc5ojh...hreads/ai-singing-music.163958/#post-16216799
 
There's a thread on AI singing here that people seem to use a Google Collab link for, with models from Huggingface or whatever Discord MEGA link they have to a character's voice model. Shit kept breaking on my end and couldn't get it to work for some reason but maybe you'll have better luck if you wanna mess around with it. https://kiwifarmsaaf4t2h7gc3dfc5ojh...hreads/ai-singing-music.163958/#post-16216799
I don't want them to sing. I want them to talk.
 
So seeing as how Elevenlabs paywalled their shit and there's been some progress in other AI voice cloning programs, I'll probably repurpose this thread to just be AI Voice Cloning general and not just Elevanlabs anymore.

In any case I'd like to inform you all about RVC, the "newest" voice cloning program available.
Document detailing how to make AI Covers (info from videos above but in text form): https://docs.google.com/document/d/13_l1bd1Osgz7qlAZn-zhklCbHpVRk6bYOuAuB78qmsE/edit?usp=sharing
Document detailing how to train your own AI Voice: https://docs.google.com/document/d/13ebnzmeEBc6uzYCMt-QVFQk-whVrK4zw8k7_Lw3Bv_A/edit?usp=sharing

Google Collab link to use when generating AI Voices: https://colab.research.google.com/drive/1Gj6UTf2gicndUW_tVheVhTXIIYpFTYc7?usp=sharing
Google Collab link to use when training new voices: https://colab.research.google.com/drive/1TU-kkQWVf-PLO_hSa2QCMZS1XF5xVHqs?usp=sharing

You can find an archive of a lot of voice models from huggingface here (though you'll probably find a lot more on their discord server unfortunately): https://huggingface.co/QuickWick/Music-AI-Voices/tree/main

Their main discord server where you can find more models and find some more help if you need to: https://discord.gg/J2Dces6v

Audio demos below:
Hitler is very excited for Christmas this year:


DSP alternative reaction after getting caught jackking live:


Vib-Haliation:


Alex Jones dropping fire this year:


Linkara finally coming out of the closet:


Parappa's gonna come out with a 3rd game with this one:


Mario "Time To Take A Piss" Guy has an announcement to make:


That one universe where Roiland said Rick and Morty got into a healthy relationship with each other:


Hazbin/Helluvafags are gonna cream themselves knowing they can dub coomcomics with these tools:
 
How do you just type the words and get them to say stuff? Or is it just doing it off audio?
 
How do you just type the words and get them to say stuff? Or is it just doing it off audio?
It does it off audio, tho if you want to use TTS for characters to say things you could probably use the free version of Elevanlabs (or TortoiseTTS) download that as an audio file and pass that through RVC for it to say off whatever character you want it to say
 

Frank Sinatra - ‘Five Nights at Freddy’s’​


 
  • Horrifying
  • Like
Reactions: Mr.Miyagi and Vecr
A full on meltdown is happening because of a persona va Screenshot_20230706_202742_Twitter.jpg


Screenshot_20230706_203006_YouTube.jpg
Screenshot_20230706_at_165134_Futaba_Sakura_sings_Welcome_to_the_Internet_Bo_Burnham_AI_Cover.png

Screenshot_20230706_203044_YouTube.jpg
Kek
 
Last edited:
Back