Geodesic
Specific Selection of FFT Amplitudes from Audio Sports and News Broadcasting for Classification Purposes
Journal of Graph Algorithms and Applications, Tome 11 (2007) no. 1, pp. 277-307.

Voir la notice de l'article provenant de la source Journal of Graph Algorythms and Applications website

In this paper we investigate the problem of classification between sports and news broadcasting. We detect and classify files that consist of speech and music or background noise (news broadcasting), and speech and a noisy background (sports broadcasting). More specifically, this study investigates feature extraction and training and classification procedures. We compare the Average Magnitude Difference Function (AMDF) method, which we consider more robust to background noise, with a novel proposed method. This method uses several spectral audio features which may be considered as specific semantic information. We base the extraction of these features on the theory of computational geometry using an Onion Algorithm (OA). We tested the classification procedure as well as the learning ability of the two methods using a Learning Vector Quantizer One (LVQ1) neural network. The results of the experiment showed that the OA method has a faster learning procedure, which we characterise as an accurate feature extraction method for several audio cases. 
@article{JGAA_2007_11_1_a11,
     author = {Marios Poulos and George Bokos and Nikolaos Kanellopoulos and Sozon Papavlasopoulos and Markos Avlonitis},
     title = {Specific {Selection} of {FFT} {Amplitudes} from {Audio} {Sports} and {News} {Broadcasting} for {Classification} {Purposes}},
     journal = {Journal of Graph Algorithms and Applications},
     pages = {277--307},
     publisher = {mathdoc},
     volume = {11},
     number = {1},
     year = {2007},
     doi = {10.7155/jgaa.00146},
     language = {en},
     url = {http://geodesic.mathdoc.fr/articles/10.7155/jgaa.00146/}
}
TY  - JOUR
AU  - Marios Poulos
AU  - George Bokos
AU  - Nikolaos Kanellopoulos
AU  - Sozon Papavlasopoulos
AU  - Markos Avlonitis
TI  - Specific Selection of FFT Amplitudes from Audio Sports and News Broadcasting for Classification Purposes
JO  - Journal of Graph Algorithms and Applications
PY  - 2007
SP  - 277
EP  - 307
VL  - 11
IS  - 1
PB  - mathdoc
UR  - http://geodesic.mathdoc.fr/articles/10.7155/jgaa.00146/
DO  - 10.7155/jgaa.00146
LA  - en
ID  - JGAA_2007_11_1_a11
ER  - 
%0 Journal Article
%A Marios Poulos
%A George Bokos
%A Nikolaos Kanellopoulos
%A Sozon Papavlasopoulos
%A Markos Avlonitis
%T Specific Selection of FFT Amplitudes from Audio Sports and News Broadcasting for Classification Purposes
%J Journal of Graph Algorithms and Applications
%D 2007
%P 277-307
%V 11
%N 1
%I mathdoc
%U http://geodesic.mathdoc.fr/articles/10.7155/jgaa.00146/
%R 10.7155/jgaa.00146
%G en
%F JGAA_2007_11_1_a11
Marios Poulos; George Bokos; Nikolaos Kanellopoulos; Sozon Papavlasopoulos; Markos Avlonitis. Specific Selection of FFT Amplitudes from Audio Sports and News Broadcasting for Classification Purposes. Journal of Graph Algorithms and Applications, Tome 11 (2007) no. 1, pp. 277-307. doi : 10.7155/jgaa.00146. http://geodesic.mathdoc.fr/articles/10.7155/jgaa.00146/

Cité par Sources :