OpenVoice V2

OpenVoice V2 is a versatile instant voice cloning model from MyShell and MIT researchers. It clones a speaker's tone color from a short reference clip and can synthesize speech in multiple languages — English, Spanish, French, Chinese, Japanese and Korean are natively supported — including zero-shot cross-lingual cloning for languages not in the massive-speaker training set.\n\nV2 improves on the original with a new training strategy for better audio quality and adds flexible voice-style control over emotion, accent, rhythm, pauses and intonation, independent of the reference speaker's own style. Since April 2024 both V1 and V2 are released under the MIT license for free commercial use.\n\nThe model has powered MyShell's instant voice cloning feature since 2023 and has been used tens of millions of times, making it one of the most battle-tested open voice-cloning systems available.