I was wondering if there were some tutorials on how to make a model run on a Raspberry Pi only when we have something to feed it?
Basically, my system will listen to a microphone for 5s when an event triggers it, and I want the 5s recording to be fed to the model afterwards, so I don’t want live inference. Is it possible? If it is, do you have an example of a tutorial I can use to understand the process?
Also, this may interest you, it provide extra functions to automatically upload your raw data with the inferred label to your project if you want to improve your model over time.