Quantcast
Share This:

Metal Science

A Neural Network's Interpretation Of THE DILLINGER ESCAPE PLAN & KRALLICE Is How Robots Listen To Metal

Posted by on December 3, 2017 at 11:07 am

Dadabots isn't your typical musician, mainly because it's not a living being at all. Instead, it's a neural network that programmers fed music into and then let it interpret what the music is. The network cranks out its own version of what it learned and tada – you've got an insane version of something insane to begin with.

I'm sure my explanation is painfully simplistic, so here's the official explanation of what's going on.

"We use a modified SampleRNN architecture to generate music in modern genres such as black metal and math rock. Unlike MIDI and symbolic models, SampleRNN generates raw audio in the time domain. This requirement becomes increasingly important in modern music styles where timbre and space are used compositionally. Long developmental compositions with rapid transitions between sections are possible by increasing the depth of the network beyond the number used for speech datasets. We are delighted by the unique characteristic artifacts of neural synthesis."

Dadabots allowed its neural network to interpret Diatoma by Krallice and Calculating Infinity by The Dillinger Escape Plan, and you can hear the results of those learnings below. As for the weird titled and artwork, "All titles were generated by a Markov chain. The album cover was created with neural style transfer."

Get updates on this story

Related Posts

COMMENTS