An aural simulation system which takes advantage of the standard Musical Instrument Digital Interface (MIDI) provides improved performance and greater flexibility in recreating the sound environment of a device being simulated. Traditional methods of generating aural cues using additive synthesis techniques have proven to be costly and often ineffective in accurately simulating complex sounds, due to a limited number of harmonic-producing elements and envelope shapes available for each sound. In the MIDI-based approach, aural cues are produced by digital sampling modules under the control of a dedicated microcomputer with built-in MIDI controllers. The sampling modules have a large amount of memory, and are capable of playing back loops of actual recorded sounds, as well as loops produced from mathematically generated waveforms or synthesizers. The MIDI interface, which was originally designed as a communications link between electronic musical instruments, is also well suited to aural simulation tasks, since it contains all the commands necessary for sound effects activation and manipulation in real-time. These commands, which include note on and off (used for enabling and disabling playback of samples), pitch bending, and channel pressure (used for amplitude changes), allow any type of aural cue to be generated. Cross-fading among multiple samples is used to reproduce dynamically varying aural cues with great accuracy. Transient and steady-state aural cues are programmed quickly and are reproduced with relatively short samples. This paper describes the hardware and software implementation of the MIDI-based aural simulation system, and how it provides a more realistic and cost-effective reproduction of the simulator sound environment. Emphasis is placed on the integration of sound analysis and sample manipulation tools into the system, and on details of managing MIDI command transfers in real-time.