Beat Tracking in Musical Recordings with Neural Networks

Abstract

Beat tracking refers to the automated detection of the beat in musical recordings. Two neural network architectures are trained and evaluated for this task.

Description

The beat tracking process is divided into four stages: audio preprocessing, network training, peak-picking and beat detection. In the practical part, two network architectures (CNN and RNN) are trained and evaluated.

Citation

Collections

Endorsement

Review

Supplemented By

Referenced By