Awesome! It's really cool that it uses public domain images as the inspiration, too. And even cooler that you link to the places where you got those public domain images.
Show HN: An e-ink frame that hears birds and draws them as 1800s illustrations
161–170 of 266 posts
Re: Show HN: An e-ink frame that hears birds and draws them as 1800s illustrations
#162My granny is almost deaf and I will test the idea for her, for she loves birds
If you try it with her, I'd love to hear what she makes of it.
Re: Show HN: An e-ink frame that hears birds and draws them as 1800s illustrations
#163The underlying classifier, BirdNET, is a traditional neural network and not an LLM: https://doi.org/10.1016/j.ecoinf.2021.101236
Are there any LLMs being widely used for audio classification? I know VLMs are being used a lot in image stuff. It always seems kind of silly to me to throw everything at an LLM. I know they’re huge and can automatically handle a huge number of tasks but something in me finds it wasteful when we could be creating easily trainable, cheap to run bespoke models for a lot of stuff
There are LLMs that support audio input, similar to those with vision support.
From my testing of open weights LLMs with audio support, they basically are only trained to recognize audio as an alternative to text input, they treat audio as basically equivalent to a transcript, and can't recognize or distinguish things like music, accents, background sounds, etc.
So they're only really good for transcribing or summarizing or using audio input in place of text input for prompts, but not anything that requires distinguishing any information about the audio that would not be present in a transcript.
It can be tempting to try to use an LLM for a variety of tasks; kind of the whole thing about an LLM is that you don't have to do a separate complex training run for every task, but can just provide instructions in natural language. But it only works as far as what the training data covers, if the training basically always treated audio and a text transcript as equivalent, the model has nothing causing it to learn other relevant features of the audio. If there's enough bird call identification in the training data of an LLM, it might be able to do that, but I think multimodal training data tends to be much more limited than the text training corpus
Re: Show HN: An e-ink frame that hears birds and draws them as 1800s illustrations
#164this is very creative, really cool idea!
Re: Show HN: An e-ink frame that hears birds and draws them as 1800s illustrations
#165I've been leveraging Birdnet-2.4 as well, but also ultrasonic-pass and activity-v1 to do broader detection with an ecological targeted microphone. This means I can capture not just bird sounds but also bats and some insects, and acoustic events such as jets and cars.
Applying heterodyne transformations to the bat recordings makes them audible which is fascinating, especially when you catch them in a feeding frenzy.
Similar intent with an indoor display, although nowhere near as beautiful as OPs eink display here. I'm more aiming for a real-time ticker, but this design is wonderful. I'm quite inspired by it!
I've been working on this in the background for a few months, if anyone is interested repo with screenshots is here: https://github.com/simonjgreen/OpenObservatory
Re: Show HN: An e-ink frame that hears birds and draws them as 1800s illustrations
#166I've been leveraging Birdnet-2.4 as well, but also ultrasonic-pass and activity-v1 to do broader detection with an ecological targeted microphone. This means I can capture not just bird sounds but also bats and some insects, and acoustic events such as jets and cars. Applying heterodyne transformations to the bat recordings makes them audible which is fascinating, especially when you catch them in a feeding frenzy. S…
>I've been leveraging Birdnet-2.4 as well,
I love HN
Re: Show HN: An e-ink frame that hears birds and draws them as 1800s illustrations
#167The underlying classifier, BirdNET, is a traditional neural network and not an LLM: https://doi.org/10.1016/j.ecoinf.2021.101236
Curious if there is a similar project for dog barks.
Re: Show HN: An e-ink frame that hears birds and draws them as 1800s illustrations
#168This is the coolest thing on HN that I've seen in a minute. The whole thing is a perfect blend of ideas with the end result of something that feels... magical. Honestly this is the highest inspiration to me as a builder as I just want to create magical experiences, even if they are just small oddities.
> This is the coolest thing on HN that I've seen in a minute.
Off-topic, but what is up with the increased use of the phrase "in a minute", presumably to mean "in a long time", lately?
I've only started encountering it in the past year.
Did it get popularized by some celebrity, tv show, influencers, etc?
Re: Show HN: An e-ink frame that hears birds and draws them as 1800s illustrations
#169This is the coolest thing on HN that I've seen in a minute. The whole thing is a perfect blend of ideas with the end result of something that feels... magical. Honestly this is the highest inspiration to me as a builder as I just want to create magical experiences, even if they are just small oddities.
> This is the coolest thing on HN that I've seen in a minute. Off-topic, but what is up with the increased use of the phrase "in a minute", presumably to mean "in a long time", lately? I've only started encountering it in the past year. Did it get popularized by some celebrity, tv show, influencers, etc?
seems like you know what you're talking about 5k% increase starting in July this year
https://trends.google.com/explore?q=I%27ve%20seen%20in%20a%2...
Re: Show HN: An e-ink frame that hears birds and draws them as 1800s illustrations
#170The underlying classifier, BirdNET, is a traditional neural network and not an LLM: https://doi.org/10.1016/j.ecoinf.2021.101236
Note that while the underlying birdnet-go project started as BirdNET only, it can now use Google Perch v2, BattyBirdNET (for bats!) and other models in the future. It's a really cool project!