It wouldn't be the first time a computer program has been able to mimic the quality of improv jazz musicians. However, with funding from the Defense Advanced Research Projects Agency (DARPA) an AI is coming that will learn to play jazz and jam with the best of the best musicians.
The hero behind this new program is jazz musician-cum-computer scientist Kelland Thomas, who tells Tech Insider that while humans are given certain structures and constraints when playing jazz, so will the computer technology be able to improvise and, yes, jam with its fleshy peers.
According to Thomas, he and his team will first have to build a database of thousands of transcribed musical performances, including those by the great Louis Armstong, Miles Davis, Charlie Parker and more.
From there, after the machine has learned certain techniques, the AI will then be trained with knowledge Thomas and his group have grafted it.
This may even include a resounding tribute to "So What" from Davis.
Thomas said of his new system:
"A human musician also builds a knowledge base by practicing and by listening and by learning and studying. So the thing we're proposing to do is analogous to the way a human learns, but eventually it will be able to do this on a much larger scale. It can scour thousands of transcriptions instead of dozens or hundreds."
Thus, within five years, Thomas hopes to build a new AI that can improvise alongside a human counterpart. Following what is to be a huge event for computer science and music, the robot will be able to manipulate real time instruments and even take to the stage with other performers.
But, in the meantime, here is an idea of what is to come by Thomas with a preview from Georgia Tech PhD student Mason Bretan and his version of a "jazz robot."
