← Back to context

Comment by zkmon

15 hours ago

Finally, we solved the problem of converting bird sounds into their pictures without involving human cognition. The great problem and the need had to wait for the great, ultra expensive, data-center-driven AI technology to appear on the horizon. AI is such a good technology that continue to solve many such great problems that humans desperately needed for their survival and progress. All at the a minimal cost of some effect on the climate, which may only slightly quicken the human extinction.

Puzzlingly, no animal species requires AI, which is a mystery that need to be solved by AI.

13 comments

zkmon

Reply
iambenm  15 hours ago

Hmm. The project uses BirdNET-Go, which does inference locally on a Raspberry Pi. Then it looks up the species against a set of already existing illustrations. So in this case it seems like the "ultra expensive, data-center-driven AI" is not really involved in the core loop of this project.

  • zkmon  15 hours ago

    The local inference is a product that came out of the data-center-driven AI.

    • dansitu  15 hours ago

      This is incorrect; the local model is a small dense neural network called BirdNET that would not have been referred to "AI" when it was published in 2021. The model outputs a bird species probability distribution and the web application displays an existing image file of the most probable bird on the screen. This is a lovely example of simple, offline, and fun project using a straightforward machine learning model.

      1 reply →

    • 40four  11 hours ago

      Incorrect. As others have pointed out, BirdNet is a “traditional” neural network. The amount of compute needed to train something like this is many orders of magnitude less than an LLM. Something like this could be trained on local hardware with enough juice, or by renting a handful of GPU’s