OBJECTIVE To determine whether regular brisk walking can promote functional recovery in community-dwelling stroke patients. PATIENTS A total of 44 chronic stroke patients, recruited in Belgium and Benin, respectively European high-income and African low-income countries. METHODS This longitudinal, single-cohort, observational study with 1 intervention period and 4 time-points of assessments (2 baseline, 1 post-intervention and 1 follow-up) was structured in 3 periods: pre-intervention period (1 month), intervention period (3 months) and follow-up period (3 month). Intervention consisted of a 3 times/week group-based brisk walking programme. Primary outcome measures were ACTIVLIM-Stroke questionnaire and the 6-minute walk test (6MWT). Secondary outcome measures were the Stroke Impairment Assessment Set (SIAS), the Hospital Anxiety and Depression Scale (HADS), and the Berg Balance Scale (BBS). RESULTS All outcome measures were stable during the pre-intervention period (p ≥ 0.16). They all improved significantly after intervention (p ≤ 0.01), except the HADS (p = 0.058). However, during the follow-up period, SIAS (p = 0.002) and BBS (p = 0.001) decreased, while ACTIVLIM-Stroke, 6MWT and HADS showed no significant change (p ≥ 0.13). CONCLUSION This study suggests regular brisk walking as an effective approach to promote functional recovery in chronic stroke survivors. However, further studies are required before generalizing these results to the whole stroke population.