AI generated music & lyrics

Miquela, machine. Listening to this and reading about AI generated music and lyrics. Pretty scary. The lost tape of the 27 club and a bunch of contents comparing AI imitation of artists’ voice. it’s hard to pick the real thing and some prefer AI version… This brings lots of concerns about copyright and I imagine voice recognition is out of the window and maybe face recognition too. Love to hear what you think, please. Any opinion / insights @TheKevster? Thanks.

It is a development that’s been going on for a while starting with the use of samples and remixes. Then more and more algorithms in the compose-tools. Last years added algorithm based automated mixing (including being able to copy the overall character of an existing mix).

We’ve even had synthesized singers - but you are right they are now very much improved and can copy their voice from an existing recording. The term “AI” sounds a bit scary, what is really happening at this stage is more applied statistics than anything else. And I think we’ll see a lot of it - the streaming companies dont make money so cheap content is right up there on their wish-list.

Some artists will welcome it and allow their voices and songs to be used (for a fee) others will try to restrict.

I also think we’ll see a backlash. These days I’m more of a hobby guy but I have started to care. All synths sounds and samples are created by ourselves and the mix is algorithm-free using the good old Sony Oxford plugins rather than the fancy new stuff. It is like painting. Real watercolour and gouache is so much more rewarding than an app and a digital pen. And it is audible/visible.


Who is the young lady, is she Miquela? She’s very real, surely!

Yep, that’s Miquela, the creation of Trevor McFedries & Sara DeCou. At least she has her own identity and so does Hatsune Miku that looks manga. What’s scary is people using the voice of real artists creating new songs and work.

Fake AI-generated Frank Ocean songs sold for $13,000 by scammer .

Sir Paul McCartney is releasing a new song using AI to remove noise, enhance the music and add some of John Lennon’s AI voicing to complete the song.

Hi @davidng - I didn’t know anything about this and I have been laid up for a few days feeling a bit poorly. Will look in more detail at it and let you know if I have any thoughts…

Zoom into Miqeula’s face and eyes…it’s a good attempt, but it ain’t Human.

Wonder how many will think it is or, worryingly, won’t even care.

It had me fooled. ‘She’ looks very real.

Please take care and get well soon @TheKevster. I am sure we all have heard and seem generated imagery and voice of artists who have left. I just haven’t thought about it all that much. With AI and ever increasing computing power and data, this capability maybe accessible by everyone soon, if not already. Saddens me AI writes lyrics and some prefer the younger voice of the artist not to mention about implications on wages.

Take care @TheKevster hope you get better soon. Was hoping we could catch up next week to check some good second hand cd shops.

1 Like

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.