yovoice

yovoice runs open speech and music models on your own machine instead of a paid cloud service. It can read text aloud with built-in voices, clone a voice from a reference clip or design one from a description, dub a script or subtitle file line by line with different characters, and generate songs or instrumentals from a style prompt and lyrics. A score editor arranges parts by bar and beat, renders them with a General MIDI sound font, and exports MIDI and stems.

Models are downloaded from within the app, and an optional language-model connection can draft lyrics and arrangements. The app's code is Apache-2.0, but some supported model weights are restricted to non-commercial use. The macOS build targets Apple silicon on macOS 14 or later; Windows builds are also available.