OpenVoice is a practical real-time voice imitation technology that can mimic a speaker's voice based on short audio clips, enabling multilingual generation. It allows for fine control over emotion, accent, intonation, and pauses, with accurate timbre cloning capabilities. It achieves zero-shot cross-language voice cloning, showcasing impressive adaptability and multifunctionality. The open source project is available at: https://github.com/myshell-ai/OpenVoice. OpenVoice demonstrates zero-shot techniques for generating speech in unsupported languages, pioneering advancements in speech technology.