# Source Separation-based Data Augmentation for Improved Joint Beat and Downbeat Tracking

arXiv: https://arxiv.org/abs/2106.08703 · HTML: https://mlsanity.com/paper/2106.08703/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 5 of its 35 references that are in the map.
- First posted on arXiv 16 Jun 2021.

## Builds on (most cited first, 5 of 5)

- [madmom: a new Python Audio and Music Signal Processing Library](https://mlsanity.com/paper/1605.07008/) (2016)
- [Mixing-Specific Data Augmentation Techniques for Improved Blind Violin/Piano Source Separation](https://mlsanity.com/paper/2008.02480/) (2020)
- [Robust Downbeat Tracking Using an Ensemble of Convolutional Networks](https://mlsanity.com/paper/1605.08396/) (2016)
- [An Overview of Lead and Accompaniment Separation in Music](https://mlsanity.com/paper/1804.08300/) (2018)
- [Downbeat Tracking with Tempo-Invariant Convolutional Neural Networks](https://mlsanity.com/paper/2102.02282/) (2021)

## Built on by (most cited first, 2 of 2)

- [JukeDrummer: Conditional Beat-aware Audio-domain Drum Accompaniment Generation via Transformer VQ-VAE](https://mlsanity.com/paper/2210.06007/) (2022)
- [MARS-Sep: Multimodal-Aligned Reinforced Sound Separation](https://mlsanity.com/paper/2510.10509/) (2025)

## Abstract (arXiv)

Due to advances in deep learning, the performance of automatic beat and downbeat tracking in musical audio signals has seen great improvement in recent years. In training such deep learning based models, data augmentation has been found an important technique. However, existing data augmentation methods for this task mainly target at balancing the distribution of the training data with respect to their tempo. In this paper, we investigate another approach for data augmentation, to account for the composition of the training data in terms of the percussive and non-percussive sound sources. Specifically, we propose to employ a blind drum separation model to segregate the drum and non-drum sounds from each training audio signal, filtering out training signals that are drumless, and then use the obtained drum and non-drum stems to augment the training data. We report experiments on four completely unseen test sets, validating the effectiveness of the proposed method, and accordingly the importance of drum sound composition in the training data for beat and downbeat tracking.