In this master thesis, a voice encryption system was programmed as a real-time software application. The idea was based on a previous graduation thesis and is meant to show an example of the possibilities of signal processing for educational use. The application uses a frequency scrambling technique on an audio signal taken from the computer microphone input and plays it scrambled back to the speakers, or other way around to descramble the signal. The basic idea is to use a frequency channel decomposition through digital signal filtering and down-sampling. Reassembling these sub-bands in a different order, a scrambled voice signal is the result. In addition, the scrambling process is embedded into a “software framework” which allows the use of an audio data stream from and to a sound card through different APIs so a computer platform real-time application can be accomplished.