1 . Artificial Intelligence Develops an Ear for Birdsong

We can learn a lot from nature if we listen to it more — and scientists around the world are trying to do just that. From mountain peaks to ocean depths, biologists are planting audio recorders to eavesdrop (窃听) on the whistles and songs of whales, elephants, bats and especially birds. This summer, for example, over 2,000 electronic ears will record the sound scape of California’s Sierra Nevada mountain range.     1    

“Audio data is a real treasure because it contains vast amounts of information,” says ecologist Connor Wood, a Cornell University postdoctoral researcher, who is leading the Sierra Nevada project. “We just need to think creatively about how to share and access that information.”     2     Fortunately the latest generation of machine-learning AI systems can process thousands of hours of data in less than a day.

Stefan Kahl, a machine-learning expert at Cornell’s Center for Conservation Bioacoustics and Chemnitz University of Technology in Germany, built BirdNET, one of the most popular avian-sound-recognition systems used today. Wood’s team will rely on BirdNET to analyze the Sierra Nevada recordings.     3     The neural network then teaches itself which features can be used to associate an input (in this case, a bird’s call) with a label (the bird’s identity).

    4     Although they analyze audio much more quickly than humans, they still lag behind in filtering out overlapping sounds to home in on a signal of interest. Some researchers see this as the next problem for AI to tackle. Even the current imperfect versions, however, enable sweeping projects that would be far too time-consuming for humans to tackle alone. “As ecologists,” Wood says, “tools like BirdNET allow us to dream big.”