Objective To evaluate whether respiratory training can improve motor function, exercise endurance, and activity of daily living (ADL) in stroke patients. Methods The randomized controlled trials of the effects of respiratory training on motor function, exercise endurance, and ADL in stroke patients were searched in PubMed, Embase, Google Scholar, China National Knowledge Infrastructure, Wanfang, and VIP Database. The search date was from the establishment of each database to December 2018. The control group received routine rehabilitation, medical treatment or other interventions, and the trial group added respiratory training on that basis. Outcome measures included the Fugl-Meyer Assessment (FMA), the 6-minute walk test (6MWT), and the modified Barthel Index (BI). The literature was independently screened by two investigators according to the inclusion and exclusion criteria, and the quality of the included articles was evaluated using the Physiotherapy Evidence Database scale and Cochrane Library systematic review criteria. Statistical analysis was performed using RevMan 5.3 software. Results A total of 11 articles with 741 stroke patients were included. Meta-analysis showed that in the trial group the exercise endurance [mean difference (MD)=41.50 m, 95% confidence interval (CI) (7.63, 75.37) m, P=0.02], ADL [MD=9.97, 95%CI (3.99, 15.96), P=0.001], and motor function [MD=8.00, 95%CI (1.29, 14.70), P=0.02] were improved compared with those in the control group. Subgroup analysis showed that after 8-10 weeks of intervention, BI of the trial group was higher than that of the control group [MD=25.37, 95%CI (16.49, 34.25), P<0.000 01]; after 8 weeks and 12 weeks to 3 months of intervention, FMA of the trial group was higher than that of the control group [ after 8 weeks: MD=20.40, 95%CI (9.72, 31.08), P=0.000 2; after 12 weeks to 3 months: MD=6.18, 95%CI (3.57, 8.79), P<0.000 01]. Conclusions The results of this study showed that respiratory training can improve exercise tolerance, ADL, and motor function in stroke patients. In consideration of the limited number of included articles as well as the heterogeneity among included articles in the current study, and the lack of long-term follow-up period, further studies could use more optimized respiratory training programs to conduct high-quality researches with bigger sample sizes.