Unable to use GPU in Automatic speech recognition using Whisper and OpenVINO with Generate API.

[VaibMittal7]

I tested the notebook for Automatic speech recognition using Whisper and OpenVINO with Generate API, I observed that it unable to use GPU. I want to run model on GPU.

Is there something I'm missing? Please assist me with this. Thank you.

Notebook link : https://docs.openvino.ai/2024/notebooks/whisper-asr-genai-with-output.html

[brmarkus]

Can you provide more details, please?
Which HW are you running on, which operating system, which version of Python do you use?
Did you encounter error messages?
Which models and which conversions have you selected in the drop-down menus before reaching the inference section in the Jupyter notebook?

[VaibMittal7]

Processor series : Intel Core Ultra 7 165H
CPU processor : Intel Core Ultra 7 165H
Python 3.10
Windows
Whisper tiny model

[brmarkus]

Do you get errors, or empty, or wrong results?

[VaibMittal7]

I checked on task manager, GPU was not used when model was running.

[brmarkus]

Can you try using longer samples (e.g. minutes instead of seconds)?

[VaibMittal7]

Ok, let me try

[VaibMittal7]

I tried using longer samples but I'm still facing same issue.