$ cat /etc/cookies.conf
We use cookies to understand how people use this site.
Analytics cookies help us improve your experience.
They are off by default. Nothing tracks you until you say so.
$ select cookie_preferences
Members-Only
Recent Talks & Demos are for members only
You must be an AI Tinkerers active member to view these talks and demos.
This demo shows surgically patching Microsoft's VibeVoice Realtime to remove CUDA dependencies and run TTS natively on Mac Metal with low latency.
Microsoft released VibeVoice for NVIDIA, but I wanted it on Mac. In this demo, I’ll show how I used Cursor to rip out the CUDA requirements and force the model to run on Metal. The result? Real-time, local TTS on an M4 Pro with zero cloud dependencies.
Loading recent emails...