Making fake instrument noises from small clips of real instrument noises: https://erlj.notion.site/Neural-Ins...-few-samples-2cf41d8b630842ee8c7eb55036a1bfd6
I feel like I'm missing. Apart from seeing if you can do it, what's the advantage of training a synthesizer to mimic an instrument from a very small number of samples, when simply giving it a larger number of samples isn't that hard?