How Humans & Machines Classify Music

Date TBA

As music services increasingly rely on algorithms and machine learning to deliver personalized recommendations to listeners, there is a growing need for machines to understand how to accurately classify music. For example, a listener can now simply ask their device to “play some Indie Rock”, and in doing so, they expect to hear Indie Rock music. But what are they really expecting here? What is “Indie Rock”? What isn’t “Indie Rock”? How can we expect machines to accurately understand musical genres when these labels are not always clear to us humans? In this talk, Evan Paul will discuss the challenges that both humans and machines face when attempting to classify music and how these challenges will shape the future of personalized music recommendations.



Programming descriptions are generated by participants and do not necessarily reflect the opinions of SXSW.

Programming descriptions are generated by participants and do not necessarily reflect the opinions of SXSW.

Evan Paul

Pandora

Details

Primary Entry:
Music Badge
Platinum Badge
Artist Wristband
Secondary Entry:
Film Badge
Interactive Badge
Format:

Focus15

Type:

Session

Track:

Making & Marketing Music presented by Playbook Hub

Level:

Beginner