DiffRhythm: Fast End-to-End Full-Length Song Generation with Latent Diffusion

Re:DiffRhythm: Fast End-to-End Full-Length Song Generation w... by jononor on Tuesday March 04, 2025 @01:46PM (#43258498)

I agree with your overall description of creation. But I do not agree that generative models are something else entirely. They are tools, and while their affordances do influence what people do with it, in the end the responsibility is on the creator. You can make "soulless shit" or "thoughtful commentary" or anything else you put your mind to, by using these tools in combination with all the existing ones.

Models that are oriented around one-shot, text-only direction are pretty limiting in creative flow. This will hopefully continue to improve.

[ Parent | on HN ]

Re:DiffRhythm: Fast End-to-End Full-Length Song Generation w... by jimbokun on Wednesday March 05, 2025 @02:20PM (#43271002)

> in the end the responsibility is on the creator.

Bullshit. These "tools" replace the creator, who now has no meaningful input to what gets created.

The composer is the AI. Not the person who spent 10 seconds typing a prompt.

[ Parent | on HN ]

Re:DiffRhythm: Fast End-to-End Full-Length Song Generation w... by jononor on Friday March 07, 2025 @07:00AM (#43289464)

To make what I consider a halfway decent song with these current easiest-to-use services (like Suno and Udiio) takes a few hours in my experience. To get there one has to work with the text, the song structure, find a decent style, and then do corrections on sections where the models goes off track. To make something that is closer to "good", I would go and re-record all the lead vocals myself, and then mix this in a DAW.

[ Parent | on HN ]