This document examines what appear to be a number of errors or otherwise questionable steps in the derivation of the model-based information rate (‘model-IR’) in Dubnov’s 2008 paper, ‘Unified view of prediction and repetition structure in audio signals with application to interest point detection’ [3]. A critique of Dubnov’s ‘Information Rate’ Samer Abdallah, Centre for Digital Music, QMUL

We propose a Bayesian method to detect change points in sequence of functional observations that are signal functions observed with noises. Since have unlimited features, it is natural think the driving underlying through an evolution process, is, different features over time but possibly at times. A change-point then viewed as cumulative effect changes many so dissimilarities before and after ...

This paper addresses the problem of change-point detection in sequences high-dimensional and heterogeneous observations, which also possess a periodic temporal structure. Due to dimensionality problem, when time between change points is order dimension model parameters, drifts underlying distribution can be misidentified as changes. To overcome this limitation, we assume that observations lie l...