A computer system for generation of auditory stimuli is described. The system produces natural-speech or software-generated stimuli for monaural, binaural, or dichotic presentation. Stimuli have been generated for experiments run both on-line and off-line.