A-BiRD uses Raspberry Pi to identify different species singing at the same time | Hacker News Reader

A-BiRD uses Raspberry Pi to identify different species singing at the same time

raspberrypi.com

37 pointsBrajeshwar7 comments

Birdnet is really cool and super easy to set up on a Pi. It's great these kids did some science with it!

When I set one up on my screen porch, I used an off-the-shelf enclosure, an inexpensive lavalier microphone (after I tried a couple other things), an Apple USB-C audio adapter, and a cheapo USB-C to USB-A adapter.

Merlin Bird ID is an app that runs the same BirdNet model locally on your phone. https://merlin.allaboutbirds.org/

now I have an other good reason to buy a raspberry pi or dive into ML. aarg. That's dope, even if analysis is done at home it's pleasing to see young people doing projects like that.

No, it doesn't. It uses RPi to record and upload singing.

> A microphone array is hooked up to a Raspberry Pi which is programmed to listen to ambient sounds, record bird songs, and show audio curves. The Pi then sends that data to BirdNET Sound ID at the Cornell Lab of Ornithology for analysis.

birdnet runs on device https://github.com/mcguirepr89/BirdNET-Pi?tab=readme-ov-file

I recently upgraded mine from a Pi 3 to a 4. Its pretty impressive, even with some pretty rubbish hardware. A fluffy wind muffler for the mic is a good idea and nowhere near as exciting as they ought to be, with a name like that!

The second link in the article goes to https://www.raspberrypi.com/news/classify-birds-acoustically... which says "It uses a USB sound card to pick up bird sounds, and classifies them locally using a pre-trained machine learning model."

But the next link https://sarsef.org/wp-content/uploads/2023/02/Finnegans-Proj... says "The computer is programmed to listen to ambient sounds, record bird songs, show audio curves, and send all data to BirdNET at the Cornell Lab of Ornithology for analysis in order to determine what kind of bird is present in a specific location".

Reply on news.ycombinator.com