VoiceStudio

An open-source AI voice studio that clones voices, dubs videos and creates audiobooks on your own hardware, with no account required.

Alternative to:

VoiceStudio is a local AI voice studio for creators who need speech generation and audio production without uploading their recordings. It combines voice creation, video dubbing and longer spoken projects in one application.

Voice cloning and design: Clone a voice from a three-second recording, describe a new voice in text, or choose a ready-made voice from the gallery.
Video dubbing: Generate translated speech for videos while preserving the timing of the original dialogue.
Stories and audiobooks: Assign voices to different roles in a script or turn a book into chapter-based audio.
Dictation and transcription: Dictate through a floating widget and convert recordings into searchable text.
Workflow automation: Process batch jobs, reuse recipes and watch folders, or connect other software and AI agents through a local API and MCP.

VoiceStudio runs on Windows, macOS and Linux, with Docker also available. Local workflows use your hardware and keep audio on your machine; they don't require an account. Remote services are optional, and usage analytics requires consent.

The default speech engine uses k2-fsa/OmniVoice, and you can choose other engines. The application supports multilingual voice work. It's free and open source under AGPL-3.0.