From: Mart Lubbers <mart@martlubbers.net>
Date: Thu, 2 Mar 2017 20:32:39 +0000 (+0100)
Subject: add excerpt for second dzhambazov paper
X-Git-Url: https://git.martlubbers.net/?a=commitdiff_plain;h=844805e280d5d10d0e088dec8c938c00d941b753;p=asr1617.git

add excerpt for second dzhambazov paper
---

diff --git a/asr.tex b/asr.tex
index bfbcf86..42af838 100644
--- a/asr.tex
+++ b/asr.tex
@@ -2,6 +2,8 @@
 \usepackage[nonumberlist,acronyms]{glossaries}
 \makeglossaries%
 \newacronym{HMM}{HMM}{Hidden Markov Model}
+\newacronym{GMM}{GMM}{Gaussian Mixture Models}
+\newacronym{DHMM}{DHMM}{Duration-explicit \acrlong{HMM}}
 \newacronym{HTK}{HTK}{\acrlong{HMM} Toolkit}
 \newacronym{FA}{FA}{Forced alignment}
 \newacronym{MFC}{MFC}{Mel-frequency cepstrum}
@@ -27,17 +29,18 @@ Berenzweig and Ellis use acoustic classifiers from speech recognition as a
 detector for singing lines.  They achive 80\% accuracy for forty 15 second
 exerpts. They mention people that wrote signal features that discriminate
 between speech and music. Neural net
-\glspl{HMM}.\cite{berenzweig_locating_2001}.
+\glspl{HMM}~\cite{berenzweig_locating_2001}.
 
 In 2014 Dzhambazov et al.\ applied state of the art segmentation methods to
 polyphonic turkish music, this might be interesting to use for heavy metal.
 They mention Fujihara(2011) to have a similar \gls{FA} system. This method uses
 phone level segmentation, first 12 \gls{MFCC}s. They first do vocal/non-vocal
 detection, then melody extraction, then alignment. They compare results with
-Mesaros \& Virtanen, 2008.
+Mesaros \& Virtanen, 2008~\cite{dzhambazov_automatic_2014}. Later they
+specialize in long syllables in a capella. They use \glspl{DHMM} with
+\glspl{GMM} and show that adding knowledge increases alignment (bejing opera
+has long syllables)~\cite{dzhambazov_automatic_2016}.
 
-t\cite{dzhambazov_automatic_2014}
-t\cite{dzhambazov_automatic_2016}
 t\cite{fujihara_automatic_2006}
 t\cite{fujihara_lyricsynchronizer:_2011}
 t\cite{fujihara_three_2008}