Whispall mirror β€” SuperTonic 3 TTS

Availability mirror of SuperTonic 3 (ONNX flow-matching TTS) as used by Whispall (local, privacy-first macOS dictation & assistant app) for on-device text-to-speech.

  • Mirrored from: Supertone/supertonic-3 at revision 724fb5abbf5502583fb520898d45929e62f02c0b β€” the exact revision Whispall pins in code (supertonic.rs). Upstream main has since moved on; this mirror preserves the pinned snapshot.
  • Purpose: hosting stability β€” protects Whispall installs against upstream repo deletion.

License

The SuperTonic 3 model weights are BigScience OpenRAIL-M licensed (use-based restrictions apply and carry over to all downstream use β€” see the upstream license text included in this repo). SuperTonic code is MIT. Β© Supertone. This repo redistributes the model unmodified with all restrictions passed through.

Downloads last month
4
Inference Providers NEW
This model isn't deployed by any Inference Provider. πŸ™‹ Ask for provider support