Audio stream segregation of multi-pitch music signal based on time-space clustering using Gaussian kernel 2-dimensional model

被引：0

作者：

Kameoka, H ^{[1
]}

Nishimoto, T ^{[1
]}

Sagayama, S ^{[1
]}

机构：

[1] Univ Tokyo, Grad Sch Informat Sci & Technol, Bunkyo Ku, Tokyo 1138656, Japan

来源：

2005 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING, VOLS 1-5: SPEECH PROCESSING | 2005年

关键词：

D O I：

暂无

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

This paper describes a novel approach for audio stream segregation of multi-pitch music signal. We propose parameter-constrained time-frequency spectrum model expressing both harmonic spectral structure and temporal curve of power envelope with Gaussian kernels. MAP estimation of the model parameters using EM algorithm provides fundamental frequency, onset and offset time, spectral envelope and power envelope of every underlying audio stream. Our proposed method showed high accuracy in pitch name estimation task of several pieces of real music performance data.

引用

页码：5 / 8

页数：4