Home >> Technology

Google's Magenta explores AI's music-writing abilities

Machine learning has already been used by artificial intelligence (AI) of Google to beat a human champion at Go - a complex game - and create psychedelic images in the style of Vincent Van Gogh. The company has now unveiled a new project to explore the artistic and musical potential of machine learning.

Google's Magenta explores AI's music-writing abilities

Magenta was launched publicly on Friday by Google research scientist Douglas Eck in a post on the TensorFlow Web site. TensorFlow is developed by the Google Brain Team and it is the open source software library for machine intelligence of the company.

An MP3 sample of the song-writing ability of Magenta demonstrates a simple piano melody, which has structure and themes. It also illustrates one of the biggest challenges which have created trouble for the researchers so far while trying to get machines to create art: how to teach AI to tell stories that humans will find compelling.

"So much machine-generated music and art is good in small chunks, but lacks any sort of long-term narrative arc," Eck wrote in his post. "Alternately, some machine generated content does have long-term structure, but that structure is provided 'to' rather than learned 'by' the algorithm."

According to Eck, Google has two goals for Magenta. The first aim of the company is to advance current machine intelligence capabilities in music and art generation and the company is also aiming to create a community of artists, coders and researchers who can help Google with the first goal.

Google's ultimate plan is to open up its Magenta site on the GitHub code repository, Eck said.
"Once we have a stable set of tools and models, we'll invite external contributors to check in code to our GitHub," he said. "If you're a musician or an artist (or aspire to be one -- it's easier than you might think), we hope you'll try using these tools to make some noise or images or videos . . . or whatever you like."