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.

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

How to use Whisper on the NPU! #2691

Open
hrshy0629 opened this issue Jan 27, 2025 · 1 comment
Open

How to use Whisper on the NPU! #2691

hrshy0629 opened this issue Jan 27, 2025 · 1 comment
Assignees

Comments

@hrshy0629
Copy link

notebooks/whisper-asr-genai/whisper- asr-Genai.ipynb

When I select the device as NPU, I notice that the terminal displays err, but the CPU and GPU work fine:

[ERROR] 09:03:02.470 [vpux-compiler] Got Diagnostic at loc(fused<{name = "__module.conv1/aten::_convolution/Convolution", type = "Convolution"}>["__module.conv1/aten::_convolution/Convolution"]) : Channels count of input tensor shape and filter shape must be the same: -9223372036854775808 ! = 80
loc(fused<{name = "__module.conv1/aten::_convolution/Convolution", type = "Convolution"}>["__module.conv1/aten::_convolution/Convolution"]): error: Channels count of input tensor shape and filter shape must be the same: -9223372036854775808 ! = 80

I'm using the latest version of openvino. My PC also supports NPU! But I can't use Whisper on NPU!

@zulkifli-halim zulkifli-halim self-assigned this Jan 28, 2025
@Munesh-Intel Munesh-Intel self-assigned this Jan 29, 2025
@zulkifli-halim
Copy link

Hi @hrshy0629

I also wasn't able to run with NPU. I believe as of now NPU will be excluded. Refer here.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants