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 ($\geq$ 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.
HingeNet: A Harmonic-Aware Fine-Tuning Approach for Beat Tracking
School of Computer Science, Fudan University, Shanghai, China(复旦大学计算机科学学院);Naval Medical Center, PLA, China(中国人民解放军海军医疗中心);Graduate School of Informatics, Kyoto University, Kyoto, Japan(京都大学信息科学研究生院);Graduate School of Advanced Science and Engineering, Hiroshima University, Hiroshima, Japan(广岛大学先进科学与工程研究生院);Shanghai Key Laboratory of Intelligent Information Processing, Fudan University, Shanghai, China(上海智能信息处理重点实验室)