A Novel Approach on Speaker Gender Identification and Verification Using DWT First Level Energy and Zero Crossing

The aim of this work is to find a new criterion for determining a range of values in order to determine the gender of a speaker. The use of the Discrete Wavelet Transform (DWT) of the Daubechies db7 parent wavelet and the computation of the zero crossing energy from the first level of the DWT was followed by computation of the values of the criterion for both genders and comparison with the value of the speech basic frequency for both genders for the same sign or sentence. The standard has a limited range of values close to the basic frequency range of the same speaker through which we can determine gender. This criterion has been tested on several men and women databases with different repeated sentences for the same person or for both genders and it gives acceptable results that can be worked on.