# HG changeset patch
# User gyorgyf
# Date 1462067821 -3600
# Node ID 967b0369ba0701af6c5d8efa549b37deab3b4606
# Parent  89ad7f8945dba8fde12c681e5a9b50738bafbf25
added 4.3 refs2

diff -r 89ad7f8945db -r 967b0369ba07 musicweb.bib
--- a/musicweb.bib	Sun May 01 02:42:09 2016 +0100
+++ b/musicweb.bib	Sun May 01 02:57:01 2016 +0100
@@ -2,13 +2,23 @@
 %% http://bibdesk.sourceforge.net/
 
 
-%% Created for George Fazekas at 2016-05-01 02:41:56 +0100 
+%% Created for George Fazekas at 2016-05-01 02:56:50 +0100 
 
 
 %% Saved with string encoding Unicode (UTF-8) 
 
 
 
+@inproceedings{hershey:07,
+	Author = {Hershey, J. R. and Olsen, P. A.},
+	Booktitle = {Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
+	Date-Added = {2016-05-01 01:49:06 +0000},
+	Date-Modified = {2016-05-01 01:56:50 +0000},
+	Pages = {317-320},
+	Title = {{Approximating the Kullback-Leibler Divergence Between Gaussian Mixture Models}},
+	Volume = {4},
+	Year = {2007}}
+
 @inproceedings{casey08,
 	Author = {Michael A. Casey and Remco Veltkamp and Masataka Goto and Marc Leman and Christophe Rhodes and Malcolm Slaney},
 	Date-Added = {2016-05-01 01:40:46 +0000},
diff -r 89ad7f8945db -r 967b0369ba07 musicweb.tex
--- a/musicweb.tex	Sun May 01 02:42:09 2016 +0100
+++ b/musicweb.tex	Sun May 01 02:57:01 2016 +0100
@@ -310,13 +310,13 @@
 % for XXXX artists with a mean track count of YYY
 We obtain audio features form the AcousticBrainz\footnote{https://acousticbrainz.org/} Web service which provides audio descriptors in each category of interest. Tracks are indexed by MusicBrainz identifiers enabling unambiguous linking to artists and other relevant metadata. For each artist in our database, we retrieve features for a large collection of their tracks in the above categories, including beats-per-minute and onset rate (rhythmic), chord histograms (harmonic) and MFCC (timbral) features.
 
-For each artist, we fit a Gaussian Mixture Model (GMM) with full covariance on each set of aggregated features in each category across several tracks and compute the distances $D_{cat}$ in each feature category using Eq.\ref{eq:dist},
+For each artist, we fit a Gaussian Mixture Model (GMM) with full covariances on each set of aggregated features in each category across several tracks and compute the distances $D_{cat}$ for the selected category using Eq.\ref{eq:dist}
 %
 \begin{equation}\label{eq:dist}
 D_{cat} = d_{skl}(artist\_model_{cat}(i), artist\_model_{cat}(j)),
 \end{equation}
 %
-where $d_{skl}$ is the symmetrised Kullback-Leibler divergence obtained using a Variational Bayes approximation for mixture models described in [Hershey, Olsen 2007] and $artist\_model_{cat}(i)$ are the model parameters (GMM weighs, means and covariance matrices) for artist $i$ in the selected category. The divergences for each artist are then ranked and the top $N$ closest artists are stored.
+where $d_{skl}$ is the symmetrised Kullback-Leibler divergence obtained using a Variational Bayes approximation for mixture models\cite{hershey:07} and $artist\_model_{cat}(i)$ are the model parameters (GMM weighs, means and covariance matrices) for artist $i$ in the selected category. The divergences for each artist are then ranked and identifiers of the top $N$ closest artists are stored.
 
 
 \subsection{Similarity by mood}