4 ms·
How comes that the stable diffuse model helps here? Does the fact that it knows what an astronaut on a horse looks like have effect on the audio? Would starting
by Aardwolf 4y ago
How comes that the stable diffuse model helps here? Does the fact that it knows what an astronaut on a horse looks like have effect on the audio? Would starting the training from an empty model work too?
- neontomo 4y agoI think they only mentioned the horse to illustrate to people that they are using the same tool as what was used to generate those types of images. It's painting a picture for the uninitiated audience. From what I understood, this model is trained on spectrograms instead of horses and the like, resulting in this product.