Change mic for birds during day

[max-mapper]

I did a test with a Birdweather PUC side by side with a BattyBirdNETPi using an WAEM2 (USB-C) on a Raspi 4B with the latest BattyBirdNet, the BirdNET_GLOBAL_6K_V2.4_Model_FP16 (2023) model, and default settings. Over a 12 hour period, the PUC detected 29 species (https://app.birdweather.com/data/mPThzm4Tgj6TEEAQdEDpfn1o) and the BattyBirdNet detected 19 (https://app.birdweather.com/data/pkvugpyUmz96LjZ3Yexm5HMR). It could be differences in the software of course, but if it's microphone related, it would be nice to have an extra settings field to specify separate microphones for birds and bats, so I can use a different USB microphone during the day.

I also noticed some bird detections were European species and I'm in North America. Is the lat/lon and week being correctly passed to birdnet analyze?

[rdz-oss]

I wonder if that gets better once the microphone is recognized with its full 256k spectrum?

[lupinthief]

I haven't done any formal tests but I've observed the same. Running with an audiomoth as microphone I get many fewer detections that if just running birdnetpi with my other USB mic (sorry, not sure off the top of my head what it is). The detections are also much less plausible. I'm in the UK and get loads of North American species detected and other impossible stuff. I agree that it may be possible that the lat/lon and occurrence threshold isn't being applied properly but it would be brilliant if could switch between mics according to the daylight settings.

[EvaldoOliveira]

I have the same and have tested a lot.
The point is that audiomoth has a high noise level up to 3k hz.
This decreases a lot the bird detection precision, as this is a frequency rang that is very relevant for some birds

[max-mapper]

So the easiest way to implement this would be a second input field in the advanced settings called "Audio Card (Birds)", and then you can use the output of arecord -l to pick your card, e.g.

$ arecord -l
**** List of CAPTURE Hardware Devices ****
card 2: U2 [USB Echo Meter Touch 2], device 0: USB Audio [USB Audio]
  Subdevices: 0/1
  Subdevice #0: subdevice #0
card 3: U2 [USB Generic USB Microphone], device 0: USB Audio [USB Audio]
  Subdevices: 0/1
  Subdevice #0: subdevice #0

you would put "hw:2" into "Audio Card" and "hw:3" for "Audio Card (Birds)", then it would just use two separate variables to pass into -D "${REC_CARD}" here

BattyBirdNET-Pi/scripts/birdnet_recording.sh

Line 62 in 2f3ede7

-D "${REC_CARD}" --use-strftime \