The relation between discrimination of silent gaps and speech‐in‐noise perception was measured in 20 normal‐hearing listeners using speech‐shaped noise as both the gap markers and the noise source for speech testing. In the gap discrimination experiment, subjects compared silent gaps marked by 60 dB SPL 250‐ms noise bursts to standards of either 5, 10, 20, 50, 100, or 200 ms. The gap results were most similar to those reported by Abel [S. M. Abel, J. Acoust. Soc. Am. 52, 519–524 (1972)] as ΔT/T decreased non‐monotonically with increased gap length. In a second experiment, the California Consonant Test (CCT) was administered at 50 dB HL via CD in three conditions: quiet, +10 S/N, and 0 S/N. Results from both experiments were correlated and the association between ΔT/T and CCT scores was generally negative. Listeners who discriminated the gaps with greater acuity typically had higher speech scores. The relation was strongest for the smaller gap standards at each S/N, or when performance for any gap duration was compared to the CCT results obtained in quiet.