Improvement in safety of formant method of text dependent verification of a speaker by a new method of segmentation of speech signal

Abstract: The new method of segmentation of speech signal is introduced in this article, it improves the safety of the formant method of text dependent verification of a speaker. The new method is based on an exponent of signal singularity.

Keywords: speaker recognition, text dependent speaker verification, formant, false acceptance and rejection rate, segmentation of speech signal, singularity, holder exponent, continuous wavelet transform

For citation:
Ruchay A. N. Improvement in safety of formant method of text dependent verification of a speaker by a new method of segmentation of speech signal. Doklady Tomskogo gosudarstvennogo universiteta sistem upravleniya i radioelektroniki, 2011, no. 2(24), – p. 2. pp. 241–246.