Treatment-refractory auditory hallucinations (TRAH) in schizophrenia often do not improve with pharmacotherapy. We performed a meta-analysis of randomized, double-blind, sham-controlled clinical trials (RCTs) that systematically examined the therapeutic effects and tolerability of adjunctive active versus sham active transcranial direct current stimulation (tDCS) for auditory hallucinations as measured by the Auditory Hallucination Rating Scale (AHRS) in schizophrenia patients with TRAH. Relevant data were extracted, checked and analyzed using the Review Manager, Version 5.3 by three independent investigators. Eight double-blind RCTs covering 329 schizophrenia patients (168 in active tDCS group, 161 in sham tDCS group) were included. Although no advantage of active tDCS on auditory hallucinations [7 RCTs, n=224; standardized mean difference (SMD): -0.33 (95% confidence interval (CI): -0.71, 0.05), P=0.09; I2 =46%] was found compared to sham, subgroup analyses revealed that active tDCS with twice-daily stimulation [6 RCTs, n=198; SMD: -0.42 (95%CI: -0.82, -0.02), P=0.04; I2 =44%] and active tDCS with ≥10 stimulation sessions [6 RCTs, n=198; SMD: -0.42 (95%CI: -0.82, -0.02), P=0.04; I2 =44%] showed a significantly better therapeutic effect than sham in improving auditory hallucinations symptoms. Meta-analyses of total psychopathology and discontinuation due to any reason were not significantly different between the active and sham tDCS groups. This meta-analysis demonstrated that the effects of tDCS for auditory hallucinations symptoms were influenced by the tDCS parameters. Twice-daily stimulation and ≥10 stimulation sessions may be needed to improve auditory hallucinations symptoms in schizophrenia with TRAH.