TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Using squeezed wav2vec2 to automatically detect owl calls

77 pointsby gfysfmabout 1 year ago

4 comments

refibrillatorabout 1 year ago
Props for mentioning BirdNET as a potentially more accessible starting point for less technical folks.<p>There are a couple relative advantages of your approach that I feel are notable though:<p>Squeezed wav2vec2 (SEW) architecture leverages Transformer layers and operates directly on time series inputs. But BirdNET converts audio to a spectrogram first and then uses 2D convolution layers (ResNet-like backbone).<p>This over-representation of inputs to BirdNET implies that SEW will be much more computationally efficient for a given audio classification task (all else held equal).<p>Plus, simply using a pre-trained SEW model and then training a linear classifier on the embeddings would almost certainly produce strong baseline results. No GPU would be necessary for that.<p>P.S. Minor typo - precision and recall are confused here:<p><i>&gt; “precision” (how many of the animal calls it notices) and its “recall” (the rate at which it makes accurate predictions).</i>
评论 #40189064 未加载
sdenton4about 1 year ago
For those who want to try it a real world version of this, the BirdClef 2024 competition is currently running: <a href="https:&#x2F;&#x2F;www.kaggle.com&#x2F;competitions&#x2F;birdclef-2024" rel="nofollow">https:&#x2F;&#x2F;www.kaggle.com&#x2F;competitions&#x2F;birdclef-2024</a><p>Lots more I can say here - I&#x27;ve been working on problems in bioacoustics for years - but for now will just leave a link to some work on using bird song embeddings from last year.<p><a href="https:&#x2F;&#x2F;www.nature.com&#x2F;articles&#x2F;s41598-023-49989-z" rel="nofollow">https:&#x2F;&#x2F;www.nature.com&#x2F;articles&#x2F;s41598-023-49989-z</a>
评论 #40189408 未加载
adultSwimabout 1 year ago
For a well-polished ready-to-use version, I recommend the Merlin Bird ID app from Cornell. <a href="https:&#x2F;&#x2F;merlin.allaboutbirds.org&#x2F;" rel="nofollow">https:&#x2F;&#x2F;merlin.allaboutbirds.org&#x2F;</a>
dekhnabout 1 year ago
I record a few hours when the birds are noisy in the morning and then run it through birdnet. It&#x27;s integrated with audacity so you can just pull up a wav and its annotations and listen through for the various species. I could imagine automating this and using it to track migration patterns, etc, around your house.