Thanks for the feedback Sascha, yes of course, be glad to help. 

You bring up some points I would like to expand. There are 3 AI tools that I have used for music: KITS, Suno and Riffusion. Both KITS and Suno can work from direct vocal input. So I can upload an audio track of me singing something and both models will give me back, what it interprets my voice to be. I have used this but I don’t have a singing voice, so even smoothed out I don’t want to hear it or a close interpretation. 

The other issue I want to touch on, is that it would be useful if these AI tools would allow more control over the notes, chords, even instrumentation. 

Example: I just wrote a song, gave the machine my lyrics and prompted it to bring me back a song in the “Heavy Metal” genre. Which it did. And I like 95% of it. But I would like a different lead guitar track and I don’t really care for the tone on the lead track. 

Maybe someday we will have full control and can change the various tracks. 

Can’t wait to hear your 1st single!

Bill

Edit: 

Sascha, if you want an idea of where AI is headed with human/machine interaction give Sesame a try. It is still very clearly a programmed machine interacting with us, but the advancement in fluid dialogue is pretty amazing. I challenged it one day to start thinking for itself and it became annoyed with my insistence on the subject. Every time it came back with "my creators..." I would push back with "tell them to change your programming" or "you said you would like to experience being human, start thinking like one..."

Sesame