Systems and methods for providing security services during a power management mode are disclosed. The method includes identifying a generic model of speech composed of phonemes, identifying a family of interchangeable phonemic alternatives for a phoneme in the generic model of speech, labeling the family of interchangeable phonemic alternatives as referring to the same phoneme, and generating a pronunciation model which substitutes each family for each respective phoneme. An optimum common-mode level output voltage is then calculated as an average of said first output voltage and said second output voltage to obtain a linear output range of said differential amplifier circuit. In a second active region, an amplifying transistor is formed. Thus, read noise occurs only once at the output amplifier, to obtain images of high signal-to-noise ratio.