To evaluate the effect of music therapy on cognitive dysfunction after stroke. Methods: Randomized controlled trials studying the effects of music therapy for poststroke cognitive dysfunction were retrieved from Chinese and English electronic databases, and meta-analysis was performed using RevMan 5.4 statistical software. Results: A total of 15 articles with 1 229 patients were included. The meta-analysis results revealed that for the Montreal Cognitive Assessment (MoCA), the pooled effect size was MD=2.91, 95% CI (2.57, 3.25), P<0.00001 for studies published before 2019, and MD= 4.62, 95% CI (4.23, 5.02), P< 0.00001 for those published in or after 2019. Significant improvements were also found in the pooled Mini-Mental State Examination (MMSE) [MD= 4.49, 95% CI (0.81, 8.18), P<0.00001], Stroop Color and Word Test [MD=5.17, 95%CI (2.99, 7.35), P<0.00001], Hamilton Anxiety Scale (HAMA) [MD=－4.55, 95% CI (－5.16, －3.93), P< 0.00001], and Hamilton Depression Scale (HAMD) [MD=－ 6.02, 95% CI (－ 6.84, － 5.20), P<0.00001]. Conclusion: Music therapy can significantly improve cognitive dysfunction and emotional state in stroke survivors.