Generative spectrogram factorization models for polyphonic piano transcription

Authors:
Paul H. Peeling;A. Taylan Cemgil;Simon J. Godsill
Affiliations:
Signal Processing and Communications Laboratory, Engineering Department, Cambridge University, Cambridge, UK;Department of Computer Engineering, Boǧaziçi University, Bebek, Istanbul, Turkey;Signal Processing and Communications Laboratory, Engineering Department, Cambridge University, Cambridge, UK
Venue:
IEEE Transactions on Audio, Speech, and Language Processing
Year:
2010

Citing 10
Cited 0

A view of the EM algorithm that justifies incremental, sparse, and other variants

Learning in graphical models
An Introduction to Variational Methods for Graphical Models

Machine Learning
Monte Carlo Statistical Methods (Springer Texts in Statistics)

Monte Carlo Statistical Methods (Springer Texts in Statistics)
A discriminative model for polyphonic piano transcription

EURASIP Journal on Applied Signal Processing
Efficient Bayesian inference for harmonic models via adaptive posterior factorization

Neurocomputing
Nonnegative matrix factorization with the itakura-saito divergence: With application to music analysis

Neural Computation
Conjugate gamma Markov random fields for modelling nonstationary sources

ICA'07 Proceedings of the 7th international conference on Independent component analysis and signal separation
A generative model for music transcription

IEEE Transactions on Audio, Speech, and Language Processing
A connectionist approach to automatic transcription of polyphonic piano music

IEEE Transactions on Multimedia
Unsupervised analysis of polyphonic music by sparse coding

IEEE Transactions on Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

We introduce a framework for probabilistic generative models of time-frequency coefficients of audio signals, using a matrix factorization parametrization to jointly model spectral characteristics such as harmonicity and temporal activations and excitations. The models represent the observed data as the superposition of statistically independent sources, and we consider variance-based models used in source separation and intensity-based models for non-negative matrix factorization. We derive a generalized expectation-maximization algorithm for inferring the parameters of the model and then adapt this algorithm for the task of polyphonic transcription of music using labeled training data. The performance of the system is compared to that of existing discriminative and model-based approaches on a dataset of solo piano music.