AI Helping to Decode Sperm Whale Language

Researchers studying sperm whale communication have made significant strides in uncovering sophisticated structures similar to those found in human language. The discovery is thanks in part to AI and is offering a glimpse into their mysterious underwater world.

For many decades, sperm whales have exhibited complex behaviors and communication patterns that have long puzzled scientists. “They’re slow swimmers,” says Kirsten Young, a marine scientist at the University of Exeter. “Squid, on the other hand, are fast. How can [sperm whales] catch squid if they can only move at 3 knots [5.5 km/h or 3.5mph]?”

The answer remains elusive, as much of the whales’ foraging and hunting occurs at depths beyond the reach of sunlight and is difficult to monitor due to depths.

They can reach depths of over 3 km (10,000 ft). All while holding their breath for up to two hours. During these dives, they communicate using rhythmic sequences of clicks called codas. These codas are now known to be more complex than previously thought.

Coda Type Surprise

Researchers at the Cetacean Translation Initiative (Ceti) have identified 156 distinct coda types, far exceeding the 21 types initially believed to exist.

David Gruber, lead and founder of Ceti and professor of biology at the City University of New York, highlights the challenges of understanding sperm whale communication. “As mammals, we share basics like eating, nursing, or sleeping with them. But it’s going to get really interesting when we try to understand areas of their world where there’s no intersection with our own.”

Advanced AI technology has played a crucial role in decoding these vocalizations. The researchers discovered that sperm whales use a “phonetic alphabet” similar to human phonemes, allowing for a vast repertoire of vocalizations.

This combinatorial coding system is a prerequisite for the “duality of patterning,” a linguistic phenomenon thought to be unique to human language. However, Pratyusha Sharma, a PhD student at MIT and lead author of the study, cautions that they have yet to find evidence of this phenomenon in whales.

AI’s contribution

The implications of these findings are exciting. This is due to the vast potential for research across the animal kingdom. There are many reasons for this, but most of all, it could help reveal insights into animal intelligence and behaviors never known.

As Gruber notes, This is now made possible thanks to AI, “We’re at base camp. This is a new place for humans to be — just give us a few years. Artificial intelligence is allowing us to see deeper into whale communication than we’ve ever seen before.”

Despite these advancements, much remains unknown. Young describes the research as an “incremental step” towards understanding these giants of the deep. She emphasizes the importance of this work for conservation efforts.

Classified as “vulnerable” by the International Union for Conservation of Nature, sperm whales face threats such as climate change, ocean noise pollution, and ship strikes. But much like the research this week from the University of Michigan, the research could go a long way toward better understanding between humans and other living organisms.

Originally posted on OpenDataScience.com

Read more data science articles on OpenDataScience.com, including tutorials and guides from beginner to advanced levels! Subscribe to our weekly newsletter here and receive the latest news every Thursday. You can also get data science training on-demand wherever you are with our Ai+ Training platform. Interested in attending an ODSC event? Learn more about our upcoming events here.