Methods and systems for providing a telephone user with one or more of a plurality of possible voice scripts are disclosed. The voice scripts may be comprised of one or more individual voice modules which are dynamically selected for presentation to the user based on information known about the user. The information known about the user may include the user's identity, information obtained from the user on one or more previous calls, and/or demographic information obtained from third party sources. The user's identity may include the user's ANI, gender, and/or other identity data. The user's gender may be determined by analyzing the tonality of the user's voice.