The problem is these devices don’t have the hardware to process input locally — it’s all sent back to their respective clouds for processing.
I believe Siri on newer phones can do some processing entirely local, but it’s not the norm.
Did you check out the Loki Docker plugin for the daemon? That worked like a charm for me.
Promtail will grab host level logs as well.
DM if you’re comfortable with Ansible; I have the whole stack (host + Docker services) automated and can share.
Much appreciated — I think the rack mounted desktop GPU approach is best for now. Another commenter suggested we should see better options in 1-2 years and I strongly suspect they’re correct.
I totally agree on the Coral TPUs. Great for Frigate, but not much else. I’ve got 2x of the USB ones cranking on half a dozen 4K stream - works wonderfully.
And I agree in theory these Nanos should be great for all sorts of stuff, but nothing supports them. Everything I’ve seen is custom one offs outside of DeepStake (though CodeProject.AI purports there’s someone now working on a Nano port).
Sounding like a decent gaming GPU and a 2-3U box is the ticket here.
I just recently saw this pop up in my feeds. Gonna give it a go when I get a chance. Looks very cool!
If you use Firefox, you can host your own sync server.