Among them [15–17] propose a low complexity time-scale expansion algorithm that preserve the speaker identity by expanding the time scale of only the voiced segments of speech signals only by a constant factor of 2, keeping the unvoiced and silence segments unchanged