Magenta is a project that lead by the Google Brain Team to explore how machines with computational abilities can generate music. Magenta encompasses two goals. It’s first a research project to advance the state-of-the art in music, video, image and text generation. So much has been done with machine learning to understand content—for example speech recognition and translation. Different with other artists utilize computation to generate music, songs and musics of Magenta was solely generated by computer intelligence with machine learning models. The first creation song from Magenta is a simple melody that primed with four notes. The Magenta algorithm is given four notes (C,C,G,G) to begin with, and it come up with the original melody. I personally find it very impressive that how artificial intelligence with deep learning models can already accomplished some simply content generation and creativity, which is a stark contrast to the stereotypical perspective that machine will only be capable of non-creative/systematic tasks.
There is some demo from Magenta: