Real-time beat tracking with undergraduates: a report on the 2017 Signal Processing Cup
In my talk I will give an overview of the recent Signal Processing Cup on Real-Time Beat Tracking which took place at the ICASSP 2017 conference. The aim of the SP Cup was for teams of undergraduate students to build a real-time beat tracking algorithm on an embedded device (such as a Raspberry Pi) and then to develop a creative application which responded to the beat of the music. I would like to present the overall design of the task with a focus on pertinent aspects related to crowd-sourcing annotations as well as the implications for the evaluation of specifically real-time beat tracking algorithms. Finally I would like to raise some discussion on the nature of creative applications derived from rhythm analysis of music signals.