Joint Beat and Downbeat Tracking with Recurrent Neural Networks
Joint Beat and Downbeat Tracking with Recurrent Neural Networks
复制标题
使用循环神经网络进行联合节拍和悲观跟踪
DOI：
--
复制
发表时间：
2016
期刊：
影响因子：
--
通讯作者：
G. Widmer
中科院分区：
文献类型：
--
作者：
Sebastian Böck;Florian Krebs;G. Widmer
本文提出了一种新的联合提取音频信号中的拍和重拍的方法。直接对幅度谱图进行操作的递归神经网络用于在多个级别上对音频信号的度量结构进行建模，并且提供清楚地区分节拍和重拍的输出特征。然后，使用动态贝叶斯网络对可变长度的条进行建模，并将预测的节拍和下拍位置与全局最佳解对齐。我们发现，所提出的模型在各种不同的音乐流派和风格上实现了最先进的性能。
In this paper we present a novel method for jointly extracting beats and downbeats from audio signals. A recurrent neural network operating directly on magnitude spectro-grams is used to model the metrical structure of the audio signals at multiple levels and provides an output feature that clearly distinguishes between beats and downbeats. A dynamic Bayesian network is then used to model bars of variable length and align the predicted beat and down-beat positions to the global best solution. We find that the proposed model achieves state-of-the-art performance on a wide range of different musical genres and styles.