Various types of orthodontic appliances can lead to speech difficulties. However, speech difficulties caused by orthodontic appliances have not been sufficiently investigated by an evidence-based method. The aim of this study is to outline the scientific evidence and mechanism of the speech difficulties caused by orthodontic appliances. Randomized-controlled clinical trials (RCT), controlled clinical trials, and cohort studies focusing on the effect of orthodontic appliances on speech were included. A systematic search was conducted by an electronic search in PubMed, EMBASE, and the Cochrane Library databases, complemented by a manual search. The types of orthodontic appliances, the affected sounds, and duration period of the speech disturbances were extracted. The ROBINS-I tool was applied to evaluate the quality of non-randomized studies, and the bias of RCT was assessed based on the Cochrane Handbook for Systematic Reviews of Interventions. No meta-analyses could be performed due to the heterogeneity in the study designs and treatment modalities. Among 448 screened articles, 13 studies were included (n = 297 patients). Different types of orthodontic appliances such as fixed appliances, orthodontic retainers and palatal expanders could influence the clarity of speech. The /i/, /a/, and /e/ vowels as well as /s/, /z/, /l/, /t/, /d/, /r/, and /ʃ/ consonants could be distorted by appliances. Although most speech impairments could return to normal within weeks, speech distortion of the /s/ sound might last for more than 3 months. The low evidence level grading and heterogeneity were the two main limitations in this systematic review. Lingual fixed appliances, palatal expanders, and Hawley retainers have an evident influence on speech production. The /i/, /s/, /t/, and /d/ sounds are the primarily affected ones. The results of this systematic review should be interpreted with caution and more high-quality RCTs with larger sample sizes and longer follow-up periods are needed. The protocol for this systematic review (CRD42017056573) was registered in the International Prospective Register of Systematic Reviews (PROSPERO).