As I understand it, you're currently generating the song in Suno, extracting the vocal stem, and then adding your own parts? I think you should try the opposite approach. Record your part, 30-60 seconds, upload it to the AI, and then extend it. By the way, Udio handles rock music much better. I’ve done a few tracks like that. For electronic tracks (Tardigrade and Shorebound Saga), I used Suno for extension, while for guitar ones (The Architect of the Cosmos and Sensemakers), I went with Udio. In those cases, my part was an instrumental intro, and the AI took over, developing the music and introducing vocals.
![[#2024MAMA] G-DRAGON - 무제(Untitled, 2014)+POWER+HOME SWEET HOME+뱅뱅뱅+FANTASTIC BABY | Mnet 241123 방송](http://i.ytimg.com/vi/Ox29z5Nf1Uk/mqdefault.jpg)
Yes i will have to try that, but first i'll wait for the Suno 4 update, with better vocals i've heard 😊👍🏼
As I understand it, you're currently generating the song in Suno, extracting the vocal stem, and then adding your own parts? I think you should try the opposite approach. Record your part, 30-60 seconds, upload it to the AI, and then extend it. By the way, Udio handles rock music much better. I’ve done a few tracks like that. For electronic tracks (Tardigrade and Shorebound Saga), I used Suno for extension, while for guitar ones (The Architect of the Cosmos and Sensemakers), I went with Udio. In those cases, my part was an instrumental intro, and the AI took over, developing the music and introducing vocals.