Title :
Study of the cepstral coefficient probability density function
Author :
Tourneret, Jean-Yves ; Lacaze, Bernard ; Castanie, Francis
Author_Institution :
ENSEEIHT/GAPSE, Toulouse, France
Abstract :
Cepstral coefficients, used in pattern recognition and classification with the k-nearest-neighbor method, give far better results than classification with the centroid distance rule. This paper proposes an analysis of cepstral coefficient probability density which reveals why the k-NN rule is in many instances a necessary tool in this particular representation space
Keywords :
pattern recognition; spectral analysis; cepstral coefficient probability density function; classification; k-nearest-neighbor method; pattern recognition; representation space; Cepstral analysis; Equations; Gaussian processes; Jacobian matrices; Parameter estimation; Pattern recognition; Probability density function; Recursive estimation; Spectral analysis; Yield estimation;
Conference_Titel :
Statistical Signal and Array Processing, 1992. Conference Proceedings., IEEE Sixth SP Workshop on
Conference_Location :
Victoria, BC
Print_ISBN :
0-7803-0508-6
DOI :
10.1109/SSAP.1992.246879
