Avian Visitors • Hacker News Board

Re: Avian Visitors

#21

Post by OkayPhysicist » Mon, Jun 01, 2026, 8:42 PM UTC

ErroneousBosh wrote: post #15

tapland wrote: post #10

I like this a lot. I've been fascinated by Suzuki Toshitakas work on mapping bird sounds to syntax and want to experiment in that space too. Ive been assisting at a wild bird rehab but not until I got pet birds (released pet birds that no owner could be found for) did I realize they make these extremely faint sounds to each other that I can sometimes just barely make out when I'm right next to them but are not the ot…

This is a bit off-the-wall but maybe you could use a variant of the approach used to see tiny movements in video? There you offset two clips by a frame and "subtract" them so that anything not moving just kind of vanishes, and purposeful movement shows up as a bright line. Maybe doing something similar with a spectrogram would work? Two spectrograms, one delayed slightly with respect to the other, and subtract one fr…

Building off that idea, if the offset between the two spectrograms was based on the relative distance between the two mics (ideally 3), you could eliminate noises not coming from a certain point in space, e.g., the birdcage.

Re: Avian Visitors

#22

Post by ErroneousBosh » Tue, Jun 02, 2026, 7:42 AM UTC

OkayPhysicist wrote: post #21

ErroneousBosh wrote: post #15

Earlier quoted context omitted.

This is a bit off-the-wall but maybe you could use a variant of the approach used to see tiny movements in video? There you offset two clips by a frame and "subtract" them so that anything not moving just kind of vanishes, and purposeful movement shows up as a bright line. Maybe doing something similar with a spectrogram would work? Two spectrograms, one delayed slightly with respect to the other, and subtract one fr…

Building off that idea, if the offset between the two spectrograms was based on the relative distance between the two mics (ideally 3), you could eliminate noises not coming from a certain point in space, e.g., the birdcage.

Oh well now I'm just going to have to build something.

Re: Avian Visitors

#23

Post by jna_sh » Wed, Jun 03, 2026, 11:14 AM UTC

kiproping wrote: post #17

jna_sh wrote: post #6

Earlier quoted context omitted.

Thanks for sharing these resources and your story! I followed a very similar path, and ended up doing a biodiversity related MSc, with my dissertation being a custom classifier for poorly detected species in Príncipe. BirdNET and Perch are phenomenal achievements, but struggle in regions where, ironically, most of the world’s biodiversity is. What you’re doing for Rwandan species is so important!!

thanks, how is your work going. Any interesting papers you did, what is an area you think needs more research.

I have alas not published it yet, but I really should. How about you? For me, on the research front, I’m very interested in methods that can be sustainably applied in remote, resource constrained locations. Heavy cloud dependent workflows to adapt a huge foundation model just aren’t practical on an island that doesn’t have 24/7 electricity and sporadic connectivity.