TOOLORA / Local reading edition
OpenVoice
Transfer voice color from a short reference into newly synthesized speech; V2 natively covers Chinese, English, French, Spanish, Japanese and Korean.
Overview
OpenVoice pairs tone-color conversion with a base TTS system: extract voice features from a reference, then convert synthesized speech to that voice. V2 natively supports Chinese, English, French, Spanish, Japanese and Korean, with documented workflows using MeloTTS. Style controls include rhythm, pauses, intonation and accent. Official local guidance targets Linux developers; V1 and V2 declare MIT. Deployment also requires checkpoints and base-synthesis dependencies.
Official sources and references
- Official websitehttps://github.com/myshell-ai/OpenVoice
- Documentationhttps://github.com/myshell-ai/OpenVoice/blob/main/docs/USAGE.md
Descriptions summarize public information; consult official websites for features, pricing and regional availability.