I wouldn't say that they aren't useful for inference (there are pretty clear performance improvements even from the asahi effort you linked) - it's just that you have to convert the model ahead of time to be compatible with the ANE which is explained in the readme docs for whisper.cpp that I linked above.
I would say though that this likely excludes them from being useful for training purposes.
I would say though that this likely excludes them from being useful for training purposes.