Voice Studio

VoiceStudio (formerly OmniVoice Studio) is an open-source desktop app built for running voice cloning and related speech tasks locally on your own machine. It’s for voice creators, developers, and teams who want transcription, dubbing, dictation, and text-to-speech without relying on an external account for the core workflow.

It supports multiple speech workflows in one place:

• Voice cloning for creating voices locally
• Dubbing for adapting video audio
• Transcription & dictation for turning speech into text
• Text-to-speech for generating audio from text
• Voice design for working on voices as assets
• Voice gallery/catalog to browse available voices (language coverage depends on the engine)

VoiceStudio also includes a local API described as OpenAI-compatible, aimed at developers who want to connect voice features to their own apps. The language catalog is listed as 646 languages, with the specific set varying by engine.

You can run it on macOS, Linux, and Windows via WSL, and it’s positioned as “local. no account,” meaning you don’t need a hosted service account to use the desktop app. Support varies, and the app emphasizes local ownership of voices and projects through its Studio, Clone, Dubbing, Projects, and Local API areas.