Most AI models are just a (1) architecture (how many layers, what equations, what optimizers, etc.) and (2) parameters (weights, biases, etc.). What happens when you take two models and merge them? Sometimes, interesting things. Model merges (sometimes, “frankenmerges”) today are primarily used by hackers, not researchers or big corporations. It’s cheap, dirty, and takes a lot of trial and error.
Model Merge (Frankenmerge)
We recently added model merging to Replay and it’s been on of the biggest hit features https://www.tryreplay.io
With artist voices it’s interesting because you can select a percentage of each, and it translates fairly nicely to the style output