A robot with four arms and eight sticks can act as a human by writing and playing its own music composition. This robot is originally made in a lab at the Georgia Institute of Technology (Georgia Tech). It pieces is generated using big data of artificial intelligence (AI) and deep learning.
As reported by Engineering, researchers filled the robot an approximately 5,000 complete songs, which includes Beethoven, Beatles, Lady Gaga, Miles Davis songs and more. Shimon, the name given to the robot, has also more than 2 million motifs, riffs, as well as links to music on its system. Aside from giving the cyborg the first four seeds or measures to use as a starting point, no individuals are involved in either the writing, composition and the performance of the music.
Shimon the robot was originally developed by the director of Georgia Tech's Center for Music Technology named as Gil Weinberg. Under its initial programming, Shimon was capable of improvising music as it performed alongside humans, using an algorithm to make sure it wasn't just copying the other performers. However, as reported by Gizmodo, thanks to the efforts of the Ph.D. student Mason Bretan, the robot Shimon has become an accomplished composer. This makes sense since the machine is capable of autonomously generating the melodic, as well as the harmonic structure of a unique song.
Bretan claims that this is the first time a robot, or any kind of such machines, has used deep learning in order to compose and create an original music. Unlike its days of improvising, when Shimon played monophonically, the robot is able to also play harmonies and chords. In addition to that, it is also creating much more like a human musician, which focuses less on the next note and more on the overall structure of the composition, instead. "When we play or listen to music, we don't think about the next note and only that next note," said Bretan.