That's the e-ink monopoly for you. Though you could probably use a slightly smaller b&w eink display with a second hand raspberry 4 to shave a lot from that number.

If you have a Kobo (~ US$ 120 or even less used) you can run this using Cobalt https://github.com/BandarLabs/Cobalt/tree/beta/apps/birds

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...

It's been around in slang for a few years with a few variants, e.g. "I haven't seen her in a hot minute." I think it's just filtered into wider cultural vernacular.

I had not heard it before (or if I did I did not interpret it correctly, eg if one told me "see you in a minute" I would interpret it as "see you in a bit".

But there is this podcast discussing it in 2021, and it comes from black community slang from 70s (which is where most slang I encounter comes from). And these terms take a while to catch up usually, but it seems it was already circulating more broadly since 2000s.

https://waywordradio.org/its-been-a-minute/

Scarlett Johansson says, “See you in a minute" in Avengers: Endgame.

This phrase has been around for years, I know because it has always infuriated me.

They took a well-defined unit of time, which is relatively short, and made it mean “some unknown but very long period of time”.

So frustrating. /oldmanyellsatcloud

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

The harness that connects to a chatbot, API or voice interaction is the place to route requests to different systems. If you remember the early days of ChatGPT it explicitly said it was routing image generation to Dall-E after embellishing your request itself first.

Determining which tool to use should be a lightweight operation but I’m not expert enough to understand exactly how much lighter than a full LLM call just to recognize it needs a different tool or model.

There’s still plenty of commercial work on specialized models, for the simple reason that they’re typically much cheaper to train and run.

Many systems now use LLMs in conjunction with specialized models.

wav2vec or similar approaches are used a lot these days, which is basically BERT with audio inputs. Whether that counts as an LLM or not, I am not sure. It's a transformer architecture in any case.

People will often reach for "easily trainable, cheap" solutions when they can; the reason people reach for Transformers and LLMs is because when you throw more data at them, they get better.

Most "Show HN" posts these days are vibe-coded garbage. It is a pleasant surprise to see one that isn't!

I wanted to say that seems like a stretch but then I often find myself visualizing my best guess at the appearance, including species, of an unknown dog. It seems like size of dog and pitch of bark are negatively correlated

Yes, here's the classifier:

def get_species(): return 'Canis familiaris'

Here are the parts I bought, I'm still getting it all setup (I have my pictures displaying, working on designing the 3D printed case now, and testing it all) but according to OpenDisplay's calculator [0] I can expect 3.4 years if I refresh the screen once every 4 hours (and currently I'm doing closer to once a day to mimic my ESP32s which need deep sleep and once-a-day refresh to get even month/months of battery on a charge).

Screen: 7.5" Monochrome eInk / ePaper Display with 800x480 Pixels (https://www.seeedstudio.com/7-5-Monochrome-ePaper-Display-wi...)

BTLE board: XIAO ePaper Display Board(nRF52840) - EN05 (https://www.seeedstudio.com/XIAO-ePaper-Display-Board-nRF528...)

Battery: 3.7V 2000mAh - (https://www.amazon.com/dp/B0FR9GH966)

Picture of it all assembled (no case yet): https://cs.joshstrange.com/zJFvGGPB

The image rendered to the screen is coming from a little service I wrote that renders a book quote it pulls from BookOrbit which syncs with KOReader (annotations, progress, etc). My plan is to set up one of these one each bookshelf/series rotating quotes I highlighted from the series.

Here are 2 ESP32-based screens I have, the one on the left is a an all-in-one [1] and the one on the right I printed the case from the TRMNL DIY kit from SeeedStudio [2]: https://cs.joshstrange.com/knk4Ns2h

Right now only the BTLE one is running OpenDisplay (the other two are semi-managed by HomeAssistant ESPHome) and all the devices are in HA for controlling them, though the ESP32's all deep-sleep for saving battery so I can can't "push" stuff to them like I can the BTLE one.

[0] https://opendisplay.org/index.html#battery

[1] https://www.seeedstudio.com/reTerminal-E1001-p-6534.html

[2] https://www.seeedstudio.com/TRMNL-7-5-Inch-OG-DIY-Kit-p-6481...

I might be misunderstanding your comment but the display is run/powered by the board so all you need to do is provide power to the board. For e-ink it uses no power to "hold" an image, only to write/refresh (sorry if I'm telling you something you already know!).

So reiterate, with a 2000mAh lipo battery I should be able to drive the board _and_ the screen for over 3 years if I only refresh it at most once every 4 hours (with "push" capabilities, I don't have to deep-sleep the board to get that battery life). At least, that's what the calculator says, I _just_ got my hardware to play with and so I can't speak from experience. Unfortunately my board is not charging the battery so I need to get a replacement which will probably take a week or two to get here.

Probably BirdNet, but Google also has an equivalent model now called Perch. https://birdnet.cornell.edu/

Complete guess but I wonder if it's the Merlin project?

https://www.birds.cornell.edu/home/merlin/