TY  - JOUR
T1  - A New Robust Resonance Based Wavelet Decomposition Cepstral Features for
Phoneme Recoszgnition
AU - Al-Hassani, Ihsan AU - Assami, Abdlnaser AU - Al-Dakkak, Oumayma 
JO  - Research Journal of Applied Sciences
VL  - 14
IS  - 8
SP  - 250
EP  - 257
PY  - 2019
DA  - 2001/08/19
SN  - 1815-932x
DO  - rjasci.2019.250.257
UR  - https://makhillpublications.co/view-article.php?doi=rjasci.2019.250.257
KW  - phoneme classification
KW  -wavelet packet decomposition
KW  -speech features
KW  -ASR
KW  -Q-factor
KW  -TIMIT
AB  - Robust Automatic Speech Recognition (ASR) is a challenging task that has been an active research
subject for the last 20 years. And still results are very modest in the highly noisy environments. In this study,
we propose a new speech parameterization method based on concatenating two wavelet packet decompositions,
one decomposition using low Q-factor wavelet and another with high Q-factor wavelet, to extract speech
features suitable for ASR task in noisy conditions. Experiments on TIMIT dataset for phonemes recognition
show that the proposed wavelet-based features outperform MFCC in all noisy conditions.
ER  -