H3 Portrait Refiner MLX assets
Speech-driven local mouth refinement for AI2Apps H3 Portrait. This bundle contains MuseTalk 1.5 Q4, YuNet 2023mar converted to a portable MLX graph, and BiSeNet BF16. It contains no H3 weights, Runtime, user recordings, or portrait images. Model code is shipped separately by the AI2Apps model Package.
Sources and pinned revisions are recorded in sources.json. Original model cards and component licenses are preserved. See NOTICE.md. No blanket relicensing.
- Downloads last month
- 34
Hardware compatibility
Log In to add your hardware
Quantized
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support