[ExI] i'll bee seeing you... was: RE: old software fun (spike)

[spike]

>… On Behalf Of Anders Sandberg
…
http://www.cnel.ufl.edu/files/1425197934.pdf
… and once it has been trained those values are used as inputs to another network that is trained to classify the sounds as insects. …This might be a decent bee detector. Anders Sandberg, 

 

 

Ja, there are perhaps a dozen ways to make this happen.  I have been trying to compare habits of my two main pollinators out back: honeybees and bumblebees.  Everything else combined would be down in the buzzing noise.  Those two bugs definitely make sounds that differ from each other.  Microphones are cheap and I know how to do realtime FFTs on them, so it should be easy to recognize pollinators from their sound.  My intuition tells me that problem is easy enough that someone has already done it better than I can do now.  I just need to find that person or product.

 

spike

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20150714/238cc163/attachment.html>