Beat and downbeat tracking improved with learned tatum grid method
Explore This! Beat and Downbeat Tracking From a Learned Tatum Grid
SoundMachine Learning
Summary
Figuring out the timing of beats and downbeats in music is important for many music applications. Traditional methods treat beats individually and then use a separate system to ensure musical flow. This paper introduces a new method that first predicts finer timing points called tatums and then groups them into beats and downbeats, considering different tempo and time signature options simultaneously. The authors show their approach leads to better continuity in tracking compared to some prior methods, although it does not yet beat the very best techniques.
What this means in practice
- •For music software developers: Create more accurate beat and downbeat detection features for music editing and DJ applications using tatum-based rhythmic grouping.
- •For audio tool builders: Build rhythm-aware audio analysis tools that better handle varying tempo and time signatures in diverse music genres.
Authors
Robert Kihlborg, André Holzapfel, Jan Schlüter
Abstract
State-of-the-art beat tracking is based on neural networks that model beats and downbeats as individual events, often followed by a separate Dynamic Bayesian Network (DBN) enforcing musical consistency. We propose a novel architecture that combines deep learning with differentiable theory-based modules: Specifically, we design a system that first predicts a tatum grid, then groups tatum events to beats and downbeats, explicitly modelling multiple options for the current tempo and time signature in a bar-pointer-like fashion. Our approach improves continuity metrics compared to existing DBN-free systems and is more versatile than the de-facto standard DBN, but still falls behind the best DBN-based methods on average. Given the many design choices and hyperparameters involved, we assume we have not yet explored the full potential of this idea and publish our code and invite others to explore this further.