He explained the reason in the video Smh. For a musician you have bad listening skills. Go back and the full video.

​@@robertgibbs1308 I doubt that putting out a feeler out is a logistical nightmare. When your primary brand is musicians seeking exposure, generative audio isn't the play.

I may do that for another project with vocals in the future. But it's simply a time and money saving measure. I did almost all the work for this in a weekend on a whim with some free time I had. Even just putting a call out to my audience for musicians who are willing to get me some music for free would require sifting through 100+ applications, and it would likely take weeks to find the right person. Then weeks to get the music, and then weeks for every release. It's a pretty big tradeoff for something that largely doesn't make a big difference for instrumental music. To me this is the perfect application for this technology. I can't think of a single downside to using AI music in this way. For vocal driven it's another story because it's more obvious currently that it's AI generated. Plus i'd need content that involves a real human on screen, which I also can't do with AI generated stuff in a convincing way.

@@AndrewSouthworth Appreciate you responding Andrew. I hope you can realise that I respect the work you put into these videos while being hesitant around the application of generative audio in these experiments. Not on just a social aspect, but also how this has potential to skew the outcomes.

How so?

Thanks dude! Glad you dig it.

Yeah if you don't have the rights to use the sample in your song, meaning permission from the artist or rights holder, it wouldn't be allowed to use it in a Meta ad. You'll very likely get away with it, but I believe it'd be violating copyright law.

@@AndrewSouthworth i see so i wouldnt be able to run the ad? facebook wont start it?

No it’ll run most likely, but if you get caught you could get in legal trouble.

@@AndrewSouthworth legal trouble meaning i will lose money instead of making money? (Btw thanks for responding haha, will avaial your ad service soon once i get bigger) also planning to buy your e-book before i run ada

By legal trouble I mean a potential lawsuit from whatever record label owns the song's you're sampling. Probably wouldn't happen, but it technically could!

99% of the artists I work with make much more from streaming than from digital downloads. One artist I know with an older demographic does amazing with downloads, but their audience is mostly 40+. If you have a young audience digital downloads are basically dead, even doing CD's and Vinyl would be better than focusing on downloads for most.

@ I disagree . 15 million streams a year is about 60k at 0.004 per streams. Whereas 3000 downloads from your core fans at 20 dollars is 60k. How many artists are pulling 15 million + a year. It’s impossible without spending huge amounts of effectively promoting Spotify as a platform rather than direct to consumer.

I know waaaay more artists doing 15+ million streams per year than artists selling 3,000 albums at $20 each. I’m not saying it’s impossible, but just a lot harder than you’re making it sound. I know dozens of artists getting more than 15 million streams per year, but only 3 that make the equivalent in digital downloads. Those 3 all have demographics of 40+ as well.

@ the point is how many thousand and thousands did they need to spend on ads.

@@AndrewSouthworth Spotify should change their royalty system from stream share to user centric. There’s no point of having active listeners , super listerners. If they still don’t pay barely anything. Their whole platform is built backwards and not for the artist. It would be fair and make more sense to pay artists based on their active listeners in a centric method than stream share. Spotify emphases so much on super listeners etc but it doesn’t mean anything. In a stream share model. Current system all it does it promote passive unengaged listening. The current system makes no sense. You spend huge amounts of money on ads In hopes for the algo to push your music out. If they changed to user centric they could actually have room to push your music to thousands of people and it wouldn’t cost Spotify a thing, and then have those programmed listeners convert to active listeners.

I used Melodist, the distribution platform I started: melodist.io/

@@AndrewSouthworth Oh that is very cool! I wasn't aware even though I have already watched some of your videos. Well I gotta, say you are already doing things better than distributors who have been on the market way longer! Gonna keep your platform in mind in the future! :)

Maybe in the future. I thought about this, but it seemed like it would add a bunch more complexity and slow things down. I did most of the work for this over a weekend, adding a second person composing music would have meant it would be several weeks or months to even get an album ready - unless they had one already sitting on their hard drive. Can always have album 2 of this project be human made! Or a 2nd experiment under another name after this one is over.

@ there’s always next time. You could even make it a requirement that applicants have 2 albums worth of unreleased music ready to go or something 👍

Yeah, but I have a new / better site here! : southworth.media/

@AndrewSouthworth ok thanku