In order to test the effect of temporal envelope enhancement on speech perception, masked identification thresholds of aCa stimuli (C=p,t,k,b,d,g,f,s,ch,v,z,j) were measured in a two-interval, four-alternative forced choice procedure, using two randomly interleaved two-down/one-up, adaptive staircases. aCa stimuli were either temporally expanded or left intact. The expanded stimuli were obtained by multiplying the original signal by the original envelope at each corresponding point in time. The masker was a gated, steady speech-shaped noise. Temporal expansion was not applied to noise. Thirty normal-hearing listeners (19–28 years) and 42 hearing-impaired listeners (48–79 years) participated in this experiment. Overall, the results showed that thresholds were slightly lower when the envelope was expanded rather than left intact. Moreover, a detailed analysis of the data revealed that only one consonant was adversely affected by envelope expansion (+2 and +3 dB) in both groups. In normal-hearing listeners, a 2-, 4-, and 5-dB improvement was found for consonants f, k, and j, respectively. In hearing-impaired listeners, the results showed that identification thresholds were improved by 3.5 dB for voiced consonants. In this group, the identification thresholds of five consonants (t,b,d,g,z) were substantially lowered by expansion by 2 to 9 dB.