There's no denying that artificial intelligence is going to replace us all, ruling the world in an unstoppable takeover. Most of us think that robots are going to replace us in the workplace, but it looks as if they might reign over music too. After all, a neural network made a black metal album, and it kicks ass.
As The Outline points out, Boston research group Dadabots have used an algorithm to create a new black metal album called Coditany of Timeness.
The researchers used software that creates an emulation of whatever data it's fed. For their source material, they chopped up Krallice's 2011 album Diotima and fed it to their computer. The neural network used that data to "guess" what would happen next, and thus a new album was created.
Here's a more scientific explanation as to how the Dadabots created a black metal AI brain:
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.
Listen to Coditany of Timeness below. Hell, we've certainly heard worse black metal.
As The Outline points out, Boston research group Dadabots have used an algorithm to create a new black metal album called Coditany of Timeness.
The researchers used software that creates an emulation of whatever data it's fed. For their source material, they chopped up Krallice's 2011 album Diotima and fed it to their computer. The neural network used that data to "guess" what would happen next, and thus a new album was created.
Here's a more scientific explanation as to how the Dadabots created a black metal AI brain:
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.
Listen to Coditany of Timeness below. Hell, we've certainly heard worse black metal.
