Purpose: The effects of type of stimuli (i.e., nonspeech vs. speech), speech (i.e., natural vs. synthetic), gender of speaker and listener, speaker (i.e., self vs. other), and frequency alteration in self-produced speech on the late auditory cortical evoked potential were examined. Method: Young adult men (n = 15) and women (n = 15), all with normal hearing, participated. P1-N1-P2 components were evoked with the following stimuli: 723-Hz tone bursts; naturally produced male and female /a/ tokens; synthetic male and female /a/ tokens; an /a/ token self-produced by each participant; and the same /a/ token produced by the participant but with a shift in frequency. Results: In general, P1-N1-P2 component latencies were significantly shorter when evoked with the tonal stimulus versus speech stimuli and natural versus synthetic speech (p less than 0.05). Women had significantly shorter latencies for only the P2 component (p less than 0.05). For the tonal versus speech stimuli, P1 amplitudes were significantly smaller, and N1 and P2 amplitudes were significantly larger (p less than 0.05). There was no significant effect of gender on the P1, N1, or P2 amplitude (p greater than 0.05). Conclusion: These findings are consistent with the notion that spectrotemporal characteristics of nonspeech and speech stimuli affect P1-N1-P2 latency and amplitude components.