VoiceStudioProduct update

VoiceStudio · v0.5.0

OmniVoice-Studio is renamed VoiceStudio in v0.5.0, while the data folder, settings and Docker image paths stay the same. A new Model Catalogue gathers TTS, transcription and LLM engines in one place with device routing and install state, and engines can be switched from the status bar. Another machine's GPU can be borrowed with a join code or QR scan, and several people can share one GPU host over revocable connections. Admin actions in server mode require an API key, and the remote UI trades it for short-lived sessions so it never sits in browser storage.

Read the original

Saved . A shortened excerpt from GitHub Releases; the original publisher retains authorship.

Show all