Hey Ryan :)
Audio/MIDI reactive LoRAs was something we'd been trying to figure out with your pack for months so thanks for releasing that wf a few weeks back, hope to finally get round to testing it tonight!
We've been using Stable Audio Open in a project that we've started to release some pieces of ( MozAIk ). One of the applications of the system is DJing with generated audio in realtime.
We use Lightning 4-step in our template wf for speed as we can have a small animated loop to accompany the generated audio. Wondering if you've had much success with the ByteDance Lightning models in any of your audio reactive workflows? (We haven't in any of our tests so far!)
Also, is there a way with your pack to create .mid's from the audio / separated stems yet for use in external applications?
Cheers