A method of controlling a graphical user interface (GUI) at a wireless device is disclosed and includes storing a set of audio GUI controls at an interactive voice response server and creating an audio GUI control string that is to be communicated to the wireless device within a voice stream. The audio GUI control string corresponds to a text string that is selectably presentable at the wireless device. Further, the method can include embedding the audio GUI control string within the voice stream. Additionally, the method can include transmitting the voice stream with the embedded audio GUI control string to the wireless device.