In this paper, we propose several high-speed area-efficient recursive discrete Fourier transform (DFT)/inverse DFT (IDFT) designs adopting the module-sharing and register-splitting schemes. The proposed core architecture achieves one multiplier reduction as well as less critical period and a saving of nearly half multiplications compared with the second-order and first-order recursive DFT structures, respectively. So as to reduce the number of computation cycles, based on the new core design, we develop the area-efficient parallel and folded recursive DFT/IDFT architectures. Moreover, due to the advantages of regular and modular structure, the resulting high-speed area-efficient recursive DFT/IDFT architectures are amenable to application-specific integrated circuit (ASIC) design.