Temporal gait variability is strongly associated with motor function and falls in the context of numerous diseases. Rhythmic auditory cueing (RAC) can influence stride-to-stride time, although its effects on temporal gait variability remain unclear. Therefore, the aim of the present cross-disease study was to examine the effects of RAC on stride time variability (STV), as well as the factors affecting changes in STV during walking with RAC. Participants with post-stroke (n=12) and orthopedic disease (n=23) performed a random block design under four conditions: comfortable walking speed (CWS) and walking with RAC (RAC 0%, RAC +10%, RAC -10%). STV was measured along with co-contraction and inter-muscular coherence of the shank muscles during walking for each condition. The contributions of the muscle activity pattern and voluntary control to the change in STV between the CWS and RAC 0% conditions were examined using hierarchical multiple regression analysis. STV was significantly lower in the RAC 0% condition than in the CWS condition (p=0.03). Hierarchical multiple regression analysis revealed that the change in STV was explained by STV in the CWS condition (β=-0.36) and by changes in co-contraction (β=0.43) and inter-muscular coherence (β=0.38) during the stance phase between the CWS and RAC 0% conditions (R2=0.56, p<0.001). These findings indicate that walking training with RAC is effective in reducing gait variability and immediately improves muscle activity patterns and excessive corticospinal activity.