WaveBeat: End-to-end beat and downbeat tracking in the time domain

Steinmetz CJ, Reiss JD. WaveBeat: End-to-end beat and downbeat tracking in the time domain. In: AES Convention 151. Audio Engineering Society; 2021. Paper 655. Available from: https://aes.org/publications/elibrary-page/?id=21518

@inproceedings{Steinmetz2021_21518,
  author = {Steinmetz, Christian J. and Reiss, Joshua D.},
  title = {{WaveBeat: End-to-end beat and downbeat tracking in the time domain}},
  booktitle = {AES Convention 151},
  note = {Paper 655},
  year = {2021},
  month = oct,
  publisher = {Audio Engineering Society},
  url = {https://aes.org/publications/elibrary-page/?id=21518}
}

TY  - CPAPER
TI  - WaveBeat: End-to-end beat and downbeat tracking in the time domain
AU  - Steinmetz, Christian J.
AU  - Reiss, Joshua D.
T2  - AES Convention 151
M1  - Paper 655
PY  - 2021
DA  - 2021/10/06
UR  - https://aes.org/publications/elibrary-page/?id=21518
PB  - Audio Engineering Society
LA  - en
AB  - Deep learning approaches for beat and downbeat tracking have brought advancements. However, these approaches continue to rely on hand-crafted, subsampled spectral features as input, restricting the information available to the model. In this work, we propose WaveBeat, an end-to-end approach for joint beat and downbeat tracking operating directly on waveforms. This method forgoes engineered spectral features, and instead, produces beat and downbeat predictions directly from the waveform, the first of its kind for this task. Our model utilizes temporal convolutional networks (TCNs) operating on waveforms that achieve a very large receptive field (= 30 s) at audio sample rates in a memory efficient manner by employing rapidly growing dilation factors with fewer layers. With a straightforward data augmentation strategy, our method outperforms previous state-of-the-art methods on some datasets, while producing comparable results on others, demonstrating the potential for time domain approaches.
ER  -