This study examined the comparative effect of recasts and prompts on EFL learners' pronunciation accuracy of consonants and vowels. Eighty-nine elementary EFL learners were selected among 117 through their performance on a piloted sample KET and were randomly assigned into two experimental groups (recast, prompt), and a control group. A piloted researcher-made pronunciation pre-treatment test was administered to the three groups to measure their pronunciation accuracy before the treatment. The three groups underwent the same amount of teaching time and received the same material. The participants in the recast group received recast and those in the prompt group received prompt. However, the participants in the control group received no specific type of corrective feedback. At the end of the treatment, the participants took a pronunciation post-test parallel with the pre-treatment test. An Analysis of Covariance (ANCOVA) was run. The results indicated that both recast and prompt had a significantly positive effect on EFL learners' pronunciation accuracy of consonants and vowels. Moreover, the results showed that there was no significant difference between the effect of recast and prompt on EFL learners' pronunciation accuracy of consonants and vowels.