Translated data: The innovative technology CoMoSVC, co-developed by the University of Hong Kong and Microsoft Research Asia, can transform a person's singing voice into that of another. It achieves this through the use of a diffusion-based teacher model and self-consistency properties for voice conversion, balancing high-quality audio transformation with rapid processing speeds. Unlike traditional iterative sampling processes, CoMoSVC achieves one-step sampling, significantly accelerating processing times while maintaining high-quality conversions. This innovative technology will offer more efficient and convenient solutions for audio conversion, providing greater creative and expressive possibilities in fields such as music production.