For an initial proof of concept about sound analysis and honey bee specific sound data I’ve chosen a phenomena beekeeper talking about for ages: It is a common knowledge that “normal” colonies with a queen sounds different than queenless hives.
I have checked this hundred times and did fail a dozen times while regular hive inspections. You have sometimes heavily buzzing hives, but all is ok and the queen is there and sometimes you have the opposite facts: All sounds normal but you know there is no queen in the hive. So the human ear seems not to be the best sensor for honey bee buzzing recognition.
By counting on technology in this special case I did a small experiment - no built-in hardware in the hive, no Arduino, no I2S mic - just an ordinary mobile phone with Android and an audio recording app Sony’s Audio Recorder: https://play.google.com/store/apps/details?id=com.sonymobile.androidapp.audiorecorder&hl=de
I started recording with the phone in my hand then I placed it under the hive. The base was not solid but had a mesh, so sound recording was easy without disturbing the bees by opening the hive.
At home I used the Sonic Visualiser http://www.sonicvisualiser.org/ to visualize the sound.
I added a so called “Peak Frequency Spectogram” and tweaked the settings a bit, in the end I got this:
Colony with queen
Colony without queen
See also the thread with the automatic OSBH algorithm based sound analysis Analyze recordings from queen vs. queenless hives using "audiohealth"
I think you can see clearly differences in the pictures, have also an ear for the sound. Do you hear differences? ;-) Isn’t this interesting??!
For me a colony with queen sound like driving in a car on the Autobahn. A deep sound is there “in the back” while the second track is more confused.