Running two different real time sound classification models simultaneously

Help
#1

Question/Issue:
Is it possible to run 2 different sound classification model with 1 microphone?
Project ID:

Context/Use case:
I have two different real time sound classification in edge-impulse, how can I modify the classify.py file in order to distribute the mic sound both of the project or run repeatedly without loss much information

#2

Hi @mrksmn,

At this time, you cannot run two different ML models in the same EI project. The workaround (Linux-only) is to create two separate projects and train two separate projects. You can then run the .eim files as separate processes: Linux Python SDK | Edge Impulse Documentation

#3

But in seperate processes, microphone is shown as busy in the second process so I can not select the same microphone

#4

In your main code, can you spin up both .eim processes (e.g. as subprocesses: subprocess — Subprocess management — Python 3.12.2 documentation) and feed them data from your microphone that you capture in your main process?

#5

Thank you so much for the information, I will try it