Introducing OpenVoice, a cutting-edge voice cloning technology that can replicate a person’s voice with just a short audio clip and generate speech in various languages. This innovative approach allows for precise control over voice styles, including emotion, accent, rhythm, and intonation, while accurately preserving the tone color of the original speaker. Surprisingly, OpenVoice can even clone voices in languages not included in its training set, making it a versatile and cost-effective option compared to other commercial APIs. The technical report and source code are available online for those interested in exploring this groundbreaking technology further.
https://research.myshell.ai/open-voice