Tool not found.
Back to Home
Open-source voice cloning, voice design, video dubbing, dictation, and transcription in 646 languages. Runs locally on your own hardware by default.
VoiceStudio is an open-source desktop app for voice cloning, voice design, video dubbing with timed speech, dictation via a floating widget, transcription, and audiobook creation, supporting 646 languages. It ships with a default local speech engine (k2-fsa/OmniVoice) and supports swapping in others, with local workflows running entirely on your own hardware; remote engines are optional, and usage analytics are collected only with explicit consent. It supports NVIDIA CUDA and Apple Silicon (Metal) acceleration, runs fully on CPU where no dedicated GPU is available, and offers a local API and MCP integration for connecting it to coding agents and other tools. It's distributed as an Electron desktop app for macOS, Windows, and Linux, with Docker support, and installs via a one-command script or platform-specific release downloads. Voice cloning carries real potential for misuse, so the project asks users to clone voices only with the subject's permission, and notes that the underlying speech models carry their own licenses that should be reviewed before commercial use. VoiceStudio itself is free and open source under the GNU AGPL v3.0 License.
