Skip to content
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.

Sign up for GitHub

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

Jump to bottom

[Feature Request]: Optimize GPU with Vulkan #235

Closed
thewh1teagle opened this issue Aug 18, 2024 · 0 comments
Closed

[Feature Request]: Optimize GPU with Vulkan #235

thewh1teagle opened this issue Aug 18, 2024 · 0 comments
Assignees
@thewh1teagle

Comments

@thewh1teagle
Copy link
Owner

thewh1teagle commented Aug 18, 2024 •
edited
Loading

Describe the feature

Recently, whisper.cpp added Vulkan support. The performance on Windows is impressive, and there are fewer bugs compared to OpenCL. Transcribing a 20-second audio clip took only 14 seconds on an AMD Ryzen 5 4500U with the medium model!

With Vulkan handling our needs, we may no longer require ROCm, CUDA, OpenCL, OpenBLAS, or similar technologies, which will simplify the build process.

Once whisper-rs PR #169 and whisper-rs PR #168 are merged, I'll integrate the updates into Vibe.

Related: ggerganov/whisper.cpp#2361
@thewh1teagle thewh1teagle self-assigned this Aug 18, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@thewh1teagle thewh1teagle

Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@thewh1teagle