Comparing Probabilistic Models for Melodic Sequences thumbnail
Pause
Mute
Subtitles
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Comparing Probabilistic Models for Melodic Sequences

Published on Nov 30, 20112758 Views

Modelling the real world complexity of music is a challenge for machine learning. We address the task of modeling melodic sequences from the same music genre. We perform a comparative analysis of two

Related categories

Chapter list

Comparing Probabilistic Models for Melodic Sequences00:00
The Problem00:25
The Data02:15
Variable Length Markov Model (VMM)04:05
Dirichlet Variable Length Markov Model (Dirichlet-VMM)06:36
Time-Convolutional Restricted Boltzmann Machine (TC-RBM) - 107:35
Time-Convolutional Restricted Boltzmann Machine (TC-RBM) - 209:33
Model Evaluation10:35
Experiments: Prediction Task12:07
Experiments: Kullback-Leibler divergence between model and data statistics13:39
Experiments: Learning Musical Features with the TC-RBM14:56
Conclusions and Current Research17:26
Thank You!19:27