-
Notifications
You must be signed in to change notification settings - Fork 15
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
feat: sync whisper.cpp (enable ggml-alloc) #123
Conversation
My test results in iPhone 13 Pro Max & iPad Air 5: ggerganov/whisper.cpp#1270 (comment) |
Do you need more testing for this to move forward? Maybe I can help or something. |
I want to finish ggerganov/whisper.cpp#1293 before enable Metal. We can still sync whisper.cpp first for ggml-alloc, but GGML_USE_METAL will not be enabled for the time being. |
cpp/coreml/whisper-encoder.mm
Outdated
MLModelConfiguration *config = [[MLModelConfiguration alloc] init]; | ||
config.computeUnits = MLComputeUnitsCPUAndGPU; | ||
//config.computeUnits = MLComputeUnitsCPUAndNeuralEngine; | ||
//config.computeUnits = MLComputeUnitsAll; |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We may keep use MLComputeUnitsAll
(considering make a param for that), because the ANE performance may be better than GPU on iOS devices.
55bb054
to
1ce6701
Compare
ggerganov/whisper.cpp#1270
It can be tested by running
example/
ornpm install https://github.com/mybigday/whisper.rn#test-alloc-and-metal
.