In contrast to fitting strategies for linear amplification that have been frequently refined for listeners with different degrees of hearing loss, we know relatively little about the effects of nonlinear amplification for differing audiometric configurations. The purpose of the current study was to determine whether increases in audibility with nonlinear amplification improved speech recognition to a comparable degree for listeners with sloping sensorineural loss as for a comparison group of listeners with flat sensorineural loss. Consonant recognition was examined as a function of audibility with wide dynamic range compression amplification and with linear amplification. For linearly amplified speech, listeners with flat and sloping loss showed similar improvements in recognition given the same increases in audibility. Results for nonlinearly amplified speech indicated that the listeners with flat loss showed a greater rate of improvement as audibility increases than the listeners with sloping loss. This difference is largely due to superior performance by the listeners with sloping loss for low-audibility speech in comparison to equivalent group performance for high-audibility speech.