Score-Informed Source Separation Based on Real-Time Polyphonic Score-to-Audio Alignment and Bayesian Harmonic Model

Author

Juanjuan Cai ; Yiyun Guo ; Hui Wang ; Ying Wang

Author_Institution

Commun. Univ. of China, Beijing, China

fYear

2014

Firstpage

672

Lastpage

680

Abstract

This paper proposes a system on the basis of guidance information from music score and Bayesian harmonic model and a two-dimensional Hidden Markov (2D-HMM) states model with particle filtering to address the separation of single-channel polyphonic music source. It is showed in a large number of experiments that in recording and synthetic polyphonic music material, the informed separation method performs well in objective performance and subjective listening experience.

Keywords

Bayes methods; audio signal processing; hidden Markov models; particle filtering (numerical methods); source separation; 2D HMM state model; Bayesian harmonic model; particle filtering; real-time polyphonic score-to-audio alignment; single-channel polyphonic music score-informed source separation; synthetic polyphonic music material; two-dimensional hidden Markov state model; Bayes methods; Estimation; Harmonic analysis; Hidden Markov models; Power harmonic filters; Source separation; Vectors; Bayesian Harmonic Models; Score-informed source separation; score-to-audio alignment;

fLanguage

English

Publisher

ieee

Conference_Titel

Computational Intelligence and Communication Networks (CICN), 2014 International Conference on

Print_ISBN

978-1-4799-6928-9

Type

conf

DOI

10.1109/CICN.2014.149

Filename

7065569