Dynamic Bayesian Networks for Symbolic Polyphonic Pitch Modeling

Symbolic pitch modeling is a way of incorporating knowledge about relations between pitches into the process of analyzing musical information or signals. In this paper, we propose a family of probabilistic symbolic polyphonic pitch models, which account for both the "horizontal" and the &q...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on audio, speech, and language processing Vol. 21; no. 9; pp. 1830 - 1840
Main Authors: Raczyński, Stanisław A., Vincent, Emmanuel, Sagayama, Shigeki
Format: Journal Article
Language:English
Published: Piscataway, NJ IEEE 01.09.2013
Institute of Electrical and Electronics Engineers
Subjects:
Accuracy
Acoustics
Applied sciences
Audio signals
Bayes methods
Computer Science
Context modeling
Data models
Detection, estimation, filtering, equalization, prediction
Dynamic Bayesian Networks
Engineering Sciences
Estimation
Exact sciences and technology
Hidden Markov models
Horizontal
Information theory
Information, signal and communications theory
Interpolation
Mathematical models
multipitch analysis
Music
Pitch estimation
Signal and communications theory
Signal and Image Processing
Signal, noise
symbolic pitch modeling
Telecommunications and information theory
Training
Transaction processing
Performance evaluation
Linear interpolation
Dynamic Bayesian Networks
Probabilistic approach
Acoustic signal
Signal estimation
Entropy
Matrix factorization
Modeling
Information analysis
Accuracy
Non negative matrix
Audio signal
Linear combination
symbolic pitch modeling
multipitch analysis
Bayes network
Musical sound
Pitch(acoustics)
ISSN:1558-7916, 1558-7924
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Symbolic pitch modeling is a way of incorporating knowledge about relations between pitches into the process of analyzing musical information or signals. In this paper, we propose a family of probabilistic symbolic polyphonic pitch models, which account for both the "horizontal" and the "vertical" pitch structure. These models are formulated as linear or log-linear interpolations of up to five sub-models, each of which is responsible for modeling a different type of relation. The ability of the models to predict symbolic pitch data is evaluated in terms of their cross-entropy, and of a newly proposed "contextual cross-entropy" measure. Their performance is then measured on synthesized polyphonic audio signals in terms of the accuracy of multiple pitch estimation in combination with a Nonnegative Matrix Factorization-based acoustic model. In both experiments, the log-linear combination of at least one "vertical" (e.g., harmony) and one "horizontal" (e.g., note duration) sub-model outperformed a pitch-dependent Bernoulli prior by more than 60% in relative cross-entropy and 3% in absolute multiple pitch estimation accuracy. This work provides a proof of concept of the usefulness of model interpolation, which may be used for improved symbolic modeling of other aspects of music in the future.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:1558-7916
1558-7924
DOI:10.1109/TASL.2013.2258012