Modified method of segmentation of speech signal by continuous wavelet transform

Download article in PDF format

Authors: Ruchay A. N.

Annotation: This article describes a modified method of segmentation of the speech signal into disjoint voiced segments by the continuous wavelet transform. This method is based on an assessment of the Holder exponent using a continuous wavelet transform. The accuracy and stability of the modified method of segmentation of the speech signal are evaluated. The modified method for the segmentation of the speech signal is compared with the standard methods of segmentation.

Keywords: automatic segmentation of speech signal, singularity, holder exponent, continuous wavelet transform