-
-
Notifications
You must be signed in to change notification settings - Fork 19
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Suspiciously few detections #159
Comments
If you run birdnet-go as a system service log goes to systemd's journal, you see log messages with If you wish to see logs as they come in run |
I got it running as a service based on your advice (thanks for that). I'm not seeing a lot of logging related to analysis in the journal output, though
|
I've sometimes had the analysis part of birdnet-go hang, could this be related? I was getting the same logs as you. Randomly does it just after mid night but restarts after a restart. |
As I've been working more closely on Birdnet-Go, I've had some opportunities to be looking at the app while it's running during the day. I've noticed there are times where I'm hearing birds very close to my desk and the microphone I have connected to Birdnet-Go, which should be very clear and easy to detect, but there will be zero detections for a long time. Sometimes 10 minutes will pass with no detections, but there was constant bird activity during that time.
I know it's not apples-to-apples but, just for reference, I ran Merlin for about 45 seconds on my phone, and it had multiple detections of 9 different species. During that same time, Birdnet-Go reported 0 detections. The microphone for Birdnet-Go is in my backyard, and I was just holding Merlin up to the window, so I wouldn't expect it to get better results.
Interestingly, if I play back the clips that are captured for detection from Birdnet-Go and I hold Merlin up to my computer, it will also detect more birds than Birdnet-Go detected for the same clip, so it doesn't seem to be an audio fidelity issue or something like that.
Is there any way I can dig more into what's happening during analysis of chunks, even when it doesn't result in a detection? I see in the code there is some logging that may help with this, but I'm not sure where to see those log statements. They don't show up in the terminal, even when running with the
-d
flag, and I don't actually have abirdnet.log
file in my Birdnet-Go folder, even though it's enabled in the config file.The text was updated successfully, but these errors were encountered: