This study describes an experimental assistive listening device, with multiple stages of spectro-temporal filtering, and evaluates it using acoustic scenes presented via a loudspeaker array. Two stages of Short-Time Target Cancellation (STTC) spatially selective speech enhancement processing, implemented using both an endfire array and a broadside array, were used as spectro-temporal post-filters for a hearing aid signal processing chain. The spatial selectivity of the speech enhancement was characterized by varying the azimuth(s) of either the singular target talker or the multiple interferer takers. Speech enhancement in a range of Signal-to-Noise Ratios (SNRs) was evaluated using acoustic scenes with mixtures of five or nine concurrent talkers. There was a demonstrated benefit for the STTC post-filtering, most notably at relatively low (≤ −3 dB) SNRs.