‘Cry Wolf’: Scientists using AI to figure out what wolves are saying - Cleveland.com

Marina: Welcome to Deep Voices — I'm so glad you're here. I'm Marina, a marine biologist and science communicator, and this is the show where we dive deep into the world of whales and dolphins: their songs, their secrets, their family bonds, and the extraordinary science working to understand what they're actually saying. Marina: A warm thank you to our sponsors who make this show possible — Ada, who helps clients weave A I into their everyday workflows; Ago Consulting, pronounced "Ahh-Go," your silicon development partners from A I all the way to SoC; and Zen Semiconductor, an umbrella company with a remarkable portfolio of businesses spanning the silicon industry. We couldn't do this without them. Marina: Today's episode is a good one. We're going to talk about what whale song and birdsong might have in common with human language — and why that finding matters more than you might think. We're diving into Project CETI, the most ambitious effort ever launched to decode what sperm whales are saying to each other. And we're going to spend some time with a mysterious whale species whose habitat has only just been pieced together — after eighteen years of patient listening. Marina: Let's go. Marina: So I want to start today with sound. With the idea that sound is not just noise — it's structure. And structure, it turns out, can be a window into meaning. Marina: Researchers have been taking a close look at birdsong and whale song together — and what they're finding is genuinely exciting. Both, it seems, follow statistical patterns that are strikingly similar to the patterns we see in human language. Now, I want to be careful here, because this is not scientists saying that whales are speaking English underwater. That's not what's happening. What they're saying is something more subtle, and in some ways more profound. Marina: Human language has this property — discovered decades ago by a linguist named George Zipf — where if you rank words by how often they appear, the most common word appears roughly twice as often as the second most common, three times as often as the third, and so on. It's called Zipf's Law, and it shows up across essentially every human language ever studied. It's one of the hallmarks of structured, meaningful communication. Marina: And here's the thing. Researchers have found that both birdsong and humpback whale song follow similar statistical regularities. The units of sound — the individual elements of a song — are not random. They're organized. They have a kind of grammar. Not grammar like "I went to the store," but grammar in the deeper sense: rules about what comes after what, about hierarchy, about structure. Marina: Humpback whales in particular sing in what researchers describe as a nested, hierarchical way. They have sounds, which combine into phrases, which combine into themes, which combine into songs, which repeat in long song sessions that can last for hours. That architecture — that layering — is something we associate with complex communication in the animal world. Marina: And the new research suggests this isn't coincidence. It may be pointing to something shared — some deep principle about how complex vocal communication evolves, whether you're a whale, a bird, or a person. That's a big idea. That's a really big idea. And it's one that's sending researchers back to their recordings with fresh eyes — and fresh ears. Marina: Which brings me to something I find endlessly fascinating — and that's how a humpback whale's physical experience actually shapes the song it sings. Marina: There's a beautiful piece of research that's been getting attention recently about how a humpback's dive — literally, what the whale is doing with its body — influences the structure of its song. Humpbacks don't have vocal cords the way we do. They produce sound using structures in their larynx, and the way pressure changes with depth affects what they can produce and how. Marina: So as a male humpback descends — and these animals can dive to hundreds of meters — the acoustic properties around him change. The pressure changes. And researchers are finding that the song changes too, in ways that are tied to that physical journey through the water column. The dive shapes the song. The body and the environment are not separate from the music — they're part of it. Marina: I think about that and I find it almost poetic. The whale isn't singing in spite of the ocean. It's singing with it. The deep water is an instrument. Marina: And these songs travel. Humpback whale songs can carry for hundreds of kilometers in the right conditions. A male singing off the coast of one island can, in theory, be heard by whales near another. That reach — that extraordinary acoustic range — is part of what makes whale song one of the most remarkable phenomena in the natural world. Marina: Now, here's where I want to make a turn — because all of this beautiful complexity in whale communication has led some very determined scientists to ask: what if we could actually decode it? Marina: Enter Project CETI. CETI stands for Cetacean Translation Initiative, and it is, without exaggeration, one of the most ambitious animal communication projects ever attempted. Their target: the sperm whale. Marina: Why sperm whales? Because sperm whales communicate using a system of clicking sounds called codas — short, patterned bursts of clicks — and these codas have structure. Different groups of sperm whales use different coda repertoires, almost like dialects. Whales that live and travel together share codas. It looks, to researchers, like a rich social communication system. And crucially, it's something that A I might be able to help crack. Marina: Project CETI is deploying an extraordinary suite of technology to do this. We're talking about underwater robots — autonomous vehicles — that can track sperm whale groups in the open ocean. We're talking about hydrophone arrays, networks of underwater microphones, recording thousands of hours of whale-to-whale communication in context. And we're talking about A I models — large machine learning systems — trained to find patterns in all of that acoustic data. Marina: The goal is not just to collect recordings. The goal is to collect recordings with context. Who is talking? Who else is present? What are the whales doing? Are they feeding? Are they socializing? Are they caring for a calf? Because meaning, as any linguist will tell you, comes from context. A sound only means something in relation to when and why it was made. Marina: And sperm whales give researchers something remarkable to work with socially. These are deeply family-oriented animals. Females and calves live in multigenerational social units — grandmothers, mothers, daughters, calves, all traveling together, caring for each other's young, apparently teaching the next generation their group's vocal traditions. That's culture. That's the passing of learned behavior through social bonds across generations. Marina: Project CETI is in its early stages — they're honest about that. Decoding a non-human communication system is genuinely hard, maybe the hardest thing anyone has tried to do in animal science. But the tools they're bringing to it — the A I, the robotics, the acoustic tech — are unlike anything that's been available before. And I, for one, am watching this very closely. Marina: Now, not every breakthrough in whale science is loud or fast. Some of the most important discoveries happen quietly, over decades, by researchers who just keep listening. Marina: I want to tell you about a study that recently came out of eighteen years — eighteen years — of whale call recordings, and what those recordings revealed about one of the ocean's most mysterious species. Marina: The whale in question is Omura's whale. You may not have heard of it, and that's completely understandable — most people haven't. Omura's whale is a baleen whale, closely related to blue and fin whales, but much smaller. It was only scientifically described as a species in 2003, from specimens that had been collected decades earlier. For a long time, almost nothing was known about where these animals actually lived in the wild, what they sounded like, or how they behaved. Marina: So researchers did the painstaking thing: they deployed underwater recording equipment — hydrophones — in various locations and they listened. And they kept listening. For eighteen years, they accumulated recordings of a call type they believed was associated with Omura's whales — a low-frequency, repetitive vocalization, produced at regular intervals. Methodical. Persistent. Almost like a heartbeat in the deep. Marina: By analyzing where and when those calls occurred, cross-referencing acoustic detections with oceanographic data, researchers were able to map out something genuinely new: a picture of where Omura's whales actually spend their time. Their preferred habitat, the conditions they gravitate toward, the parts of the ocean they call home. Marina: That kind of knowledge — knowing where an animal is, reliably, across seasons — is the foundation of conservation. You can't protect what you can't find. And for a species as little-known as Omura's whale, this eighteen-year listening project is the difference between invisibility and the first outlines of a portrait. Marina: And there's something I want to say about this kind of science, because I think it deserves more credit than it gets. This isn't glamorous work. Nobody writes headlines about a researcher sitting with acoustic data on a Tuesday afternoon in year eleven of a monitoring project. But this is how real knowledge accumulates. Slowly, carefully, with tremendous patience. And then one day you look up and you realize: we know something now that we didn't know before. That's how science works at its best. Marina: I also want to mention — because it fits so perfectly here — that this approach of using sound to find and understand animals is the same fundamental logic driving efforts in other species too. Researchers are using A I to analyze wolf vocalizations — figuring out what wolves might be communicating through their howls and calls. The methodology travels. What we learn decoding one species teaches us how to listen to another. Marina: And that's a quietly hopeful thing. We're developing, collectively, a new literacy. A way of listening to the rest of the living world that we've never had before. Marina: Which brings me to my favorite kind of story — the kind that lives at the meeting point between a wild animal and a human being. Where the ocean reaches toward the shore, and something extraordinary happens. Marina: I want to talk about a humpback whale named Humphrey. Now, Humphrey became famous — genuinely famous, the kind of famous where people lined riverbanks to watch — because of what happened not once, but twice, in the San Francisco Bay Area. Marina: In 1985, a humpback whale swam under the Golden Gate Bridge and kept going. Upstream. Into fresh water, into the Sacramento River. Humpbacks are ocean animals. They are not supposed to be in rivers. And yet there was Humphrey — named by the researchers and rescue teams who scrambled to respond — swimming further and further from salt water, from the food and the environment his body needed. Marina: The rescue effort that followed was remarkable. Teams used underwater speakers to broadcast recordings of humpback feeding sounds — the kinds of vocalizations that humpbacks make when they've found food. And it worked. Slowly, Humphrey turned around. He followed the sound back toward the bay, back toward the ocean, and eventually swam back out under the Golden Gate Bridge. Tens of thousands of people watched from the banks and the bridges. It made the evening news for weeks. Marina: And then, in 1990, Humphrey came back. The same whale — identified by his unique fluke markings — swam into the bay again and got stranded on a mudflat near the town of Rio Vista. And once again, the same playbook: underwater speakers, humpback feeding sounds, a slow and patient coaxing. And once again, it worked. Humphrey made it back to the ocean. Marina: I love this story for so many reasons. But the one I keep coming back to is this: the rescue worked because the researchers understood something about how humpbacks communicate and respond to sound. They used whale language — or at least, whale-meaningful sound — to reach across the gap between species. They spoke, in some sense, to the whale. And the whale responded. Marina: That's not magic. It's biology, and careful observation, and years of accumulated knowledge about how these animals relate to sound. But it feels like something more, doesn't it? It feels like a conversation. A tentative, high-stakes conversation across a very wide divide. Marina: And I think that's what all of this — the song research, the A I projects, the eighteen-year listening studies — is ultimately reaching toward. A deeper conversation. A more honest one. One where we stop assuming that because an animal doesn't speak our language, it has nothing to say. Marina: The ocean is full of voices. We're only beginning to learn how to hear them. Marina: What a world we live in — where we can deploy underwater robots to follow sperm whales across the open ocean, train A I on their conversations, and piece together eighteen years of mysterious clicks into the first map of a species' hidden life. Where we can learn that whale song and birdsong carry echoes of the same mathematical structures that shape human language. Where we can coax a lost humpback home by playing the sounds of a meal. Marina: These aren't small things. They're telling us something important about the nature of minds in the sea — and maybe something important about the nature of minds in general. Marina: Deep Voices is brought to you by Ada, helping clients bring A I into the heart of their work; Ago Consulting — "Ahh-Go" — your partners in silicon development from A I to SoC; and Zen Semiconductor, a family of companies spanning the silicon world. Their support keeps these stories coming, and I'm genuinely grateful for it. Marina: If today's episode moved you, curious, wonder-filled you — please subscribe, share it with someone who loves the ocean, and leave a review if you have a moment. It means the world. You'll find links to all the research we talked about today in the show notes — the link is right there waiting for you. Stay curious, stay listening, and I'll see you in the deep.