WingNews logo WingNews GitHub [2]
top | item 32376943

Earbirding – How to Visualize Sounds

141 points| mellosouls | 3 years ago |earbirding.com | reply

21 comments

order
[+] adamgordonbell|3 years ago|reply
If you never seen some of what audio restoration people do with a spectrograph view in Izotope RX, you should check it out. They clean up audio using a process somewhat akin to photoshopping images. The ears are still involved, but it's a very visual process.

I suspect doing it a lot leads to some purposeful synthesia where you can hear things and know what they would look like in a spectrograph.

Here's the first example I could find: https://www.youtube.com/watch?v=XKNYYR-uUEo

[+] dekhn|3 years ago|reply
The guy who solved voice recognition at Google years ago told me that after looking at enough spectrograms with human speech he could tell exactly what was being said. "Oh, that shape- it's an 'ooo' noise."

Armed with that knowledge I started looking at more spectrograms of birds (I record birds around my house and pass through bird.net's model. Amusingly, I kind of learned eveyrthing in this article just by looking at lots of spectrograms- they are far, far easier for me to parse than any other musical representation.

I just found 'demucs' and am really enjoying doing source separation and then looking at the resulting spectrograms, then back at the original full musical spectrogram- it helps me parse out the instruments a lot better when they overlap.

There is also a product called 'Celemony DNA' which can do some editing in spectrogram space to remove individual notes.

[+] TheRealPomax|3 years ago|reply
Heck, if you don't even know that modern editors come with a "why would I look at an amplitude when the information we care about is in the spectrogram?" view, you're going to learn something pretty special.

The audio is a landscape, and anything that isn't a "global" operation (like normalizing, noise reduction, cross-fades) ranging from from just touching it up a but literal sound-from-nothing is about sculpting that landscape.

Rather than Photoshop, it's more like 3D Modeling for your audio. (while using a top-down view in RX, although some tools will let you toggle between 2d and 3d because seeing the landscape is quite valuable)

[+] monetus|3 years ago|reply
On that note, I've been told Beethoven painted his music.
[+] sdenton4|3 years ago|reply
Pieplow's books on bird song are fantastic. Each page has a tiiiiny picture of the bird up in the corner, then a big spread of spectrograms and a decent chunk of text to give the spectrograms context: eg, what are the 'typical' parts of it, vs what varies, how likely it is that subsequent vocalizations are the same/different, what to expect in variations by sex and age, and so on.

But my favorite part of the book is the introduction. In addition to just a great intro to reading spectrograms and what's up with bird vocalization physically (they have a syrinx, not a larynx), there's a lengthy section with a kind of 'phonetics' of bird song. It's great.

[+] peanut_worm|3 years ago|reply
there are quite a few bird species that are so similar that the only reliable way to tell them apart is their vocalizations. a good example is red eyed vireos and black whiskered vireos.

I am terrible with calls, this article will be very helpful for the fall migration, thanks OP.

[+] bacheaul|3 years ago|reply
After listening to some of the samples and thinking how nice they were, I've just also noticed the morning bird calls outside my home office window, which my mind usually filters out and ignores...

They're lovely!