Deep Learning for style transfer between paintings is impressive. I'm waiting for one that transfers music styles.
For example, imagine a system that generates lullabies for any given song. A Rockabye Baby generator:
A Universal Music Translation Network
http://prostheticknowledge.tumblr.com/post/174156873191
> We present a method for translating music across musical instruments, genres,and styles.
It seems I won't have to wait that long :-)
These AI attempts of mimicking human behavior, so free of real cognition & context awareness, remind me of that sweet cockatoo that mimicks his human's sounds.
The bird's sounds almost fit the occasion (anger vs. joy) but are broken and seem to lack awareness of why it is the time to swear or to laugh.
Swearing: https://www.youtube.com/watch?v=XM8aBESf8EI&t=1m
