SPEECH CODING AND NOISE REDUCTION USING ICABASED SPEECH FEATURES
JongHwan Lee 1 , HoYoung Jung 2 , TeWon Lee 3 , SooYoung Lee 1, 2 Electronics and Telecommunications Research Institute
(TEL: +82428698031, FAX: +82428698570, Email: jhlee@neuron.kaist.ac.kr)
In this paper, we have obtained efficient speech fea
tures using independent component analysis to human
speeches. When independent component analysis is ap
plied to speech signals for efficient encoding the adapted
basis vectors resemble Gaborlike features. Then only
a few active coefficients of the trained basis vectors
are sufficient for encoding the speech signals. Those
trained speech features can be used in automatic speech
recognition systems, and the proposed method gives
better recognition rates than conventional melfrequency
cepstral coefficients (MFCCs) features. Trained basis
vectors can be also applied for the removal of Gaussian
noise. Speech signal corrupted by additive white Gaus
sian noise shows much improvements on the signalto
noise ratio (SNR) after the denoising process. Then,
these denoised speech features show better recognition
performances than MFCCs features.