Native support for English, Spanish, French, Chinese, Japanese and Korean
Audio quality has been greatly improved
Easily clone any voice, speak in multiple languages, and control emotional accents
OpenVoice allows fine control of sound style, including emotion, accent, rhythm, pause and intonation, while being able to replicate the timbre of the reference speaker.
Launched OpenVoice, a versatile instant speech cloning method that only requires reference to a short audio clip of speakers to copy their voices and generate speech in multiple languages. In addition to copying the timbre of the reference speaker, OpenVoice also allows fine control of speech style, including emotion, accent, rhythm, pause and intonation. OpenVoice can also implement zero-sample cross-language speech cloning for languages not included in the massive speaker training set. OpenVoice is also highly computationally efficient, and its cost is dozens of times lower than lower-performing commercial APIs.
If you want to learn more, you can click on the link below the video.
Thank you for watching this video. If you like it, please subscribe and like it. thank
Website:https://research.myshell.ai/open-voice
Github: https://github.com/myshell-ai/OpenVoice
Thesis:https://arxiv.org/pdf/2312.01479.pdf
Online demonstration:https://www.lepton.ai/playground/openvoice
Video: