Computer Engineering and Applications ›› 2019, Vol. 55 ›› Issue (3): 115-119.DOI: 10.3778/j.issn.1002-8331.1710-0170

Previous Articles     Next Articles

Improved Melody Extraction Algorithm Based on Pitch Salience

LI Qiang, YU Fengqin   

  1. School of Internet of Things Engineering, Jiangnan University, Wuxi, Jiangsu 214122, China
  • Online:2019-02-01 Published:2019-01-24


李  强,于凤芹   

  1. 江南大学 物联网工程学院,江苏 无锡 214122

Abstract: Aiming at the mutual interference of different sound sources in polyphonic music, the pitch sequence of the same sound source is discontinuous. According to the continuity of the pitch salience and the stability of the higher harmonics, a new method of creating pitch contours is proposed that based on pitch static likelihood function and pitch salience dynamic likelihood function. Before extracting melody pitch contour, to take advantage of the timbral inconsistency of different sound sources, the Mel-frequency cepstral coefficients of the pitch contour is proposed as the timbre feature and the timbre characteristics are also calculated from the harmonic amplitudes of the pitch contour. The improved algorithm is simulated on the RECHSET music datasets, and the results show that it achieves the raw pitch estimation of 62.04% and the overall accuracy of 55.08%.

Key words: the continuity of the pitch salience, the stability of the higher harmonics, likelihood function, melody extraction

摘要: 针对复调音乐中不同声源的相互干扰导致的同一声源音高序列不连续,利用音高显著性的连续性和高次谐波的稳定性,提出基于音高静态似然性函数和音高显著性动态似然函数的创建音高轮廓方法;在提取旋律音高轮廓之前,为了利用不同声源音色的不一致性,提出计算音高轮廓的梅尔频率倒谱系数作为音色特征以及从音高轮廓的各次谐波幅度中计算音色特征。改进算法在RECHSET音乐数据集上进行仿真实验,结果表明达到了62.04%的音高估计精度和55.08%的总精度。

关键词: 音高显著性的连续性, 高次谐波的稳定性, 似然性函数, 旋律提取