The purpose of this study was to investigate the perception of transformations by varying the voice onset time (VOT) of the repeating stimulus, its repetition number, and intensity level. Fifteen subjects rated the 5‐, 25‐, and 55‐ms VOT stimuli during periods of repetitions so that transformations could be identified according to their place and voicing characteristics. The number of stimulus repetitions was 5, 32, or 95, and the sound‐pressure level of the repeating stimulus was 50, 70, or 90 dB. The major findings were as follows: (1) The 90‐dB level condition produced significantly more transformations than either the 50‐ or 70‐dB conditions for the 55‐ms stimulus only; (2) A significant difference in number of transformations was obtained for each paired comparison of the 55‐ms stimulus, but only between the 5‐and the 32‐ and 95‐ repetitions pairs for the 5 and 25 ms stimuli; and (3) Significant correlations were obtained between the voiceless and alveolar features of transformations for the 55‐ms stimulus, and between the voiced and alveolar and velar features of transformations for the 5‐ and 25‐ms stimuli. The findings support a transformation process sensitive to voicing and place cues and a peripheral component of transformations.