dancedance 4 Posted 9 hours ago Share Posted 9 hours ago Hope I don't get cancelled for using AI, but this was just a mini project to practice vocal processing/mixing and I thought I'd share It's very pre-TF-demos-core and I hope you enjoy it! 1 Quote Link to post Share on other sites More sharing options...
Addison Rae 15,344 Posted 9 hours ago Share Posted 9 hours ago love it! next can you please make her sing obsessed by addison rae (me) sitting on his lap sipping diet pepsi 1 Quote Link to post Share on other sites More sharing options...
hELXIG 41,932 Posted 8 hours ago Share Posted 8 hours ago Haha did you train your model on her The Fame voice? This sounds amazing. Did you get all the stems and convert all of the vocal layers to Gaga? That's dedication I'll be myself until they fūcking close the coffin. 1 Quote Link to post Share on other sites More sharing options...
dancedance 4 Posted 8 hours ago Author Share Posted 8 hours ago 9 minutes ago, hELXIG said: Haha did you train your model on her The Fame voice? This sounds amazing. Did you get all the stems and convert all of the vocal layers to Gaga? That's dedication Haha yes I was going for an earlyGa sound! It was challenging finding clean backing vocals for the song, but I found a karaoke version that did the job. The thing with AI vocals is that if a song has multi vocal layers it'll try to blend them all into a glitchy mess 1 Quote Link to post Share on other sites More sharing options...
hELXIG 41,932 Posted 8 hours ago Share Posted 8 hours ago (edited) 19 minutes ago, dancedance said: Haha yes I was going for an earlyGa sound! It was challenging finding clean backing vocals for the song, but I found a karaoke version that did the job. The thing with AI vocals is that if a song has multi vocal layers it'll try to blend them all into a glitchy mess Oh honey you don't have to tell me. I have a lot of experience with AI vocals. I spent literally weeks making my Gaga model and I've been playing around with it constantly. It's harder than people think. It's funny though because I can so tell that yours is FameGa because it sounds different to mine, which is trained on BTW, TFM, ARTPOP, and a touch of Joanne and Chromatica I made things like this around the release of Disease https://voca.ro/123TBtqfoTcn Edited 8 hours ago by hELXIG I'll be myself until they fūcking close the coffin. 1 Quote Link to post Share on other sites More sharing options...
dancedance 4 Posted 7 hours ago Author Share Posted 7 hours ago 8 minutes ago, hELXIG said: Oh honey you don't have to tell me. I have a lot of experience with AI vocals. I spent literally weeks making my Gaga model and I've been playing around with it constantly. It's harder than people think. It's funny though because I can so tell that yours is FameGa because it sounds different to mine, which is trained on BTW, TFM, ARTPOP, and a touch of Joanne and Chromatica I made things like this around the release of Disease https://voca.ro/123TBtqfoTcn HELP THAT AUDIO AI dubs is so fun and interesting, I could spend hours just messing around and amusing myself with it. It's very easy to to misuse it in a controversial way (Spamming streaming services for quick revenue etc) but to me it's just a fun hobby and tool to have something to practice sound mastering on. 1 Quote Link to post Share on other sites More sharing options...
hELXIG 41,932 Posted 7 hours ago Share Posted 7 hours ago 14 minutes ago, dancedance said: HELP THAT AUDIO AI dubs is so fun and interesting, I could spend hours just messing around and amusing myself with it. It's very easy to to misuse it in a controversial way (Spamming streaming services for quick revenue etc) but to me it's just a fun hobby and tool to have something to practice sound mastering on. My relationship to AI is SO love-hate. HATE it. Hate how dangerous it is and how it's ruining so much for us so quickly and the environmental impact... kinda can't stop playing with it and watching videos on it. For example just last weekend got curious and wondered what would happen if I trained an AI model on a dataset of multiple different voices instead of just one, and I created a whole new original voice it was so fascinating https://vocaroo.com/1d6AVOjehPWn I'll be myself until they fūcking close the coffin. 1 Quote Link to post Share on other sites More sharing options...
Featured Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.