I’ve been working on audio.cpp, a native C++ inference framework for audio models built on top of ggml. The framework currently has 25 model families, but I want to be precise about its state: 12 are released in the repo now and ready for normal use. I’m not counting anything still in integration or
