What if you Could you mix and match different songs from your favorite artists or create new ones yourself with their voices?
This could become a reality sooner than later, as artificial intelligence models similar to those used to create computer-generated artistic images and incorporate deepfakes into videos are increasingly being applied to music.
The use of algorithms to create music is not new. Researchers have used computer programs to generate piano scores since the 1950s, and musicians of that era such as Iannis Xenakis and Gottfried Koenig even used them to compose their own music.
What has changed are improvements in generative algorithms, which first gained popularity in 2014, along with large amounts of computing power that are increasingly changing what computers can do with music today.
OpenAI recently released a project called JukeBox, which uses the complex raw audio form to help create completely new music tracks based on a person's choice of genre, artist and lyrics. Meanwhile, tools like Amazon's AWS DeepComposer and those released by the Google Magenta project are helping to democratize developers' ability to experiment with deep learning algorithms and music.
As for commercial use, start-ups like Amper Music, which allows users to create personalized and royalty-free music, are seeing companies adopt computer-generated pieces for a series of use cases surrounding tracks background for video and they started playing record labels around with music written by AI.
As the technology and quality of computer-generated music matures, it will likely bring many changes in the media industry from individual artists to record labels for music streaming companies and submit a series of legal questions about computer-generated music rights.
Source link
