To systematically evaluate the efficacy of respiratory muscle training (RMT) on swallowing dysfunction after stroke. Methods: Relevant randomized controlled trials (RCTs) were searched in electronic databases including VIP, Wanfang Data, China National Knowledge Infrastructure (CNKI), Chinese Biomedical Literature Database (CBM), as well as PubMed, Embase, Web of Science, and the Cochrane Library. The search covered Chinese and English literature published from the inception of each database to August 2024. Meta-analysis was conducted using RevMan 5.4 software. Stata 18.0 software was employed to assess publication bias, and the GRADE system was used to evaluate the quality of evidence. Results: A total of 23 RCTs involving 1 233 patients were included. The Meta-analysis results demonstrated that, compared with conventional swallowing therapy, the experimental group showed significant improvements in clinical total effective rate [OR=5.24, 95% CI (3.41, 8.04), P< 0.001], videofluoroscopic swallowing study (VFSS) score [MD=1.51, 95% CI(1.25, 1.77), P< 0.001], Standardized Swallowing Assessment (SSA) score [MD=－ 4.91, 95% CI (－ 6.37, － 3.46), P<0.001], Penetration-Aspiration Scale (PAS) score [MD=－0.87, 95% CI (－1.35, －0.39), P<0.001], and Functional Oral Intake Scale (FOIS) score [MD=0.42, 95% CI (0.26, 0.57), P<0.001]. According to the GRADE evidence quality evaluation, the evidence levels for clinical total effective rate, VFSS, and FOIS were of moderate quality, while those for SSA and PAS were of low quality. Conclusion: Current evidence suggests that RMT can improve swallowing function in stroke patients, enhance the strength of swallowing muscles, and reduce the occurrence of aspiration and choking.