Deduce way to work around Xcode ANE compile every time you build and run

vade / OpenAI-Whisper-CoreML

OpenAI's Whisper ported to CoreML

144 stars 3 forks source link

Deduce way to work around Xcode ANE compile every time you build and run #6

Open vade opened 1 year ago

vade commented 1 year ago

Theres some stupidity with the ML Package + size of the Whisper Models, especially the larger ones that seems to not cache the compiled mlmodelc files for some reason.

This causes major development slow downs when building and running in Xcode - at least on my Xcode at 14.2

Any suggestions welcome.