recorded audio as input to the prompt?

[hemna]

I've been using suno to create some instrumental tunes from stems I create locally at home just to jam and learn. Is there a mechanism to input a raw .wav file (that I can record from my phone) as input to the prompt and have the model use that?

[gabotechs]

I'm afraid that it's not possible. I wish it was though! the technology is there, but this project did not get to implement this.

[aronsonben]

@gabotechs are you aware of any other open source music gen LLM projects? googling the phrase leads here but it seems y'all have paused development, eh? no sweat, appreciate the existence of this as it is, just looking to see if i can contribute to a community working on something anywhere.

[Nurb4000]

Ace-Step is pretty good. it supports things like that. ( tho personally i have had little luck doing it in practice )

[gabotechs]

@gabotechs are you aware of any other open source music gen LLM projects?

Not really, I lost track of all this topic since I last committed to this project. Unfortunately I cannot afford to allocate any time to this. Glad that you are finding it useful anyway!