Is it possible to use Neural computer stick 2(OpenVINO)

Archive - Mycroft Project Archive - Former Random Mycroft Category
1

Am I able to run mimic3 and or mycroft-precise using Neural computer stick 2(OpenVINO). It has built in way to convert tensor flow models to models it uses.

https://docs.openvino.ai/2021.2/omz_models_public_mozilla_deepspeech_0_8_2_mozilla_deepspeech_0_8_2.html

From the above  onnxruntime link above
OpenVINO™ Execution Provider with Onnx Runtime on Linux installed from PyPi.org come with prebuilt OpenVINO™ libs and supports flag CXX11_ABI=0. So there is no need to install OpenVINO™ separately.

To enable CX11_ABI=1 flag, build Onnx Runtime python wheel packages from source. For build instructions, please see the [BUILD page](https://onnxruntime.ai/docs/build/eps.html#openvino). OpenVINO™ Execution Provider wheels on Linux built from source will not have prebuilt OpenVINO™ libs so we must set the OpenVINO™ Environment Variable using the full installer package of OpenVINO™:

C:\ <openvino_install_directory>\setupvars.bat

Also here it looks like it is a possible combination

2

What do you want to use it for? It’s got limited model compatibility, so it’s not generally applicable to all tf uses.

3

the ONNX models on the raspberry pi to see if it would speed up response times

4

probably better off asking on the openvino forums, tbh.

5

the onnx models for mimic 3

6

See above, also maybe inquire with the onnx folks?

7

Perhaps this developer can help you out;

He has a lot of convertors and to the looks of it ,have experience with openvino, TF and ONNX.

1 Like
8

Thank you! I will check it out