Open AccessMulti‐objective optimisation of wavelet features for phoneme recognition
Author(s) -
Vignolo Leandro Daniel,
Rufiner Hugo Leonardo,
Milone Diego Humberto
Publication year - 2016
Publication title -
iet signal processing
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.384
H-Index - 42
ISSN - 1751-9683
DOI - 10.1049/iet-spr.2015.0568
Subject(s) - wavelet packet decomposition , wavelet , computer science , speech recognition , pattern recognition (psychology) , representation (politics) , noise (video) , artificial intelligence , second generation wavelet transform , lifting scheme , wavelet transform , politics , political science , law , image (mathematics)
State‐of‐the‐art speech representations provide acceptable recognition results under optimal conditions, though their performance in adverse conditions still needs to be improved. In this direction, many advances involving wavelet processing have been reported, showing significant improvements in classification performance for different kinds of signals. However, for speech signals, the problem of finding a convenient wavelet‐based representation is still an open challenge. This study proposes the use of a multi‐objective genetic algorithm for the optimisation of a wavelet‐based representation of speech. The most relevant features are selected from a complete wavelet packet decomposition in order to maximise phoneme classification performance. Classification results for English phonemes, in different noise conditions, show significant improvements compared with well‐known speech representations.