Scientists have long known that animals communicate in sophisticated ways, but deciphering those signals is slow and difficult. Prairie dogs, for example, use different yips for coyotes than for hawks, and many species combine sound with visual cues, touch, or pheromones. For hard-to-observe animals like deep-diving whales or subterranean mole rats, the challenge is even greater.
New tools are changing the picture. Hydrophones capture whale song, wearable biologgers record bat squeaks, and machine learning can now distinguish which crow made which call. Researchers at the Earth Species Project hope their NatureLM model—trained on human speech, music, and animal sounds—will eventually interpret vocalizations across species. But that promise is still theoretical.
Researchers disagree on how close we are. Con Slobodchikoff of Zoolingua argues that many animals share enough evolutionary structure that we should be able to understand them. Yossi Yovel, a neuroecologist at Tel Aviv University, worries that AI models will overemphasize acoustic similarities and lump together sounds with different meanings. As for two-way conversation, it is not clear when—or whether—that will happen.