Developed by MyShell TTS. It is capable of replicating the voice of a speaker using only a short snippet of the speaker’s audio, and then generates speech in multiple languages.
OpenVoice provides granular control over voice styles, including emotion, accent, tempo, pauses, and intonation, while replicating the timbre of a reference speaker.
Key features:
- Accurate Voice Cloning: OpenVoice can accurately clone reference voices and generate speech in multiple languages and accents.
- Flexible Vocal Style Control: Users can control the emotion and accent of the generated speech, as well as other style parameters such as tempo, pauses, and intonation.
- Zero-shot cross-language voice cloning: OpenVoice can clone voices between any language that is not included in a large-scale multilingual training set.
Website: http://research.myshell.ai/open-voice
GitHub：http://github.com/myshell-ai/OpenVoice
Online Demo: http://lepton.ai/playground/openvoice
Create your own voice bot: http://myshell.ai
Video: