Google’s Magenta Makes Music Independently

The six researcher group of Google displayed how artificial intelligence could make music. Attendees of a techno-cultural music and computation festival, Moogfest witnessed Project Magenta where the machine using Google’s TensorFlow, an open source library for neural network framework and machine learning, it made music. The priority of project Magenta will be to use procreative algorithms to produce music after which it will also craft images and music.

In the presentation a researcher played some notes which the computer picked to produce a complete new complex yet complete melody. Google announced the project would go live to public on June 1, 2016. There is a smartphone application which is in progress which would allow the users to cruise through the imaginative constructions of Project Magenta.

Earlier machines have been used to create art specifically procedurally produced art from back in the sixties. There is the Spirograph in which the geared circles with holes in them are used to make very basic patterns of computational art but even for something simple like this there was some type of coding by a person with an idea or expectancy of a final work was required. 350 years later scientists created Al which could master and create Rembrandts (a famous painter) brush strokes.

Project Magenta is however a completely an innovative project where the machine independent and void of any human touches. It makes music itself; in Moogfest it took just five basic notes to create a complex tune. Google has earlier presented what the dreams of learning machine looked like, where they made machines generate identifiable images over frequent repetitions.

Google is assured about Magenta to improve in future and that it could be used to create music for the mass, imaginably the Google could use it to make an album.

  Related Posts