Sperm Whales are about 15 million years older than humans. Their highly developed, massive brains allow them to plan, communicate, and interact socially for decades at a time.
Sperm whales are one of the smartest animals on the planet. Not only do they have a morse code-like language we cannot understand, but sperm whales are very important for the environment, as their massive 30-plus-ton bodies store carbons playing an important role in cooling the planet. Clicks, or codas, are the rhythmic series that sperm whales produce when they “speak.”
Researchers on the Caribbean island of Dominica have begun a five-year effort to decipher sperm whales’ Morse code-like communications. It is part of the Cetacean Translation Initiative (CETI). This program seeks to gain a deeper understanding of the ocean’s most intelligent predators and protect their habitats from further human interference. CETI is estimated to be the largest interspecies communication effort in history, involving scientists from 16 different research agencies. “We’re dealing with a completely unknown form of communication, and gathering the data can be challenging,” said Begus, an assistant professor of linguistics in UC Berkeley’s Division of Social Sciences.
The engineering team at CETI is developing a drone system capable of automatically tagging whales and recording their vocalizations and other parameters. CETI’s engineering team is building a drone system that will automatically tag whales and record their vocalizations and other parameters, such as orientation, speed, and temperature. The giant brains of sperm whales, which are about 15 million years older than humans, allow them to plan, communicate, and interact socially for decades at a time. Besides storing carbon, their silver bodies also serve as secondary cooling systems for the planet.
In sperm whales, clans and families are organized by matrilineality, each group having a distinct dialect. An assortment of organs in their skulls known as spermaceti produces their powerful clicking sounds. In contrast to dogs’ bark, whales’ sound patterns are learned, not innate, according to Begus. Noise pollution, climate change, commercial fishing, and military activities such as underwater explosions and sonar technologies threaten their habitats in deep oceans. In the event that sperm whales are allowed to die naturally, they sink to the ocean floor, carrying with them carbon that would have gone into the atmosphere had they been hunted and brought ashore. Furthermore, whale droppings increase phytoplankton growth, which captures about 40% of carbon dioxide emissions.
“If we get to know sperm whales better by learning their communication and the full scope of their cognitive and social life, it’s harder for us as a species to treat them like non-sentient beings and destroy them,” Begus said. UC Berkeley researchers Begus and Goldwasser are using artificial intelligence, including deepfake technology, to identify linguistic patterns in recordings of sperm whale clicks. “In the case of translation of one human language to another, a “rosetta stone” is often available, which makes this a so-called supervised language translation problem. “Even when such examples are not known, at the very least, we have a good sense of what may be the general topics and context in which conversations are taking place to detect when a proposed translation is nonsensical,” said Goldwasser, who, at UC Berkeley, is director of the Simons Institute for the Theory of Computing.
“For CETI,” she added, “we need to significantly extend the theory and practice of unsupervised language translation, where no correct translation examples are given, to a setting where our prior knowledge of what the whales may be communicating about is limited, and we can’t run controlled experiments. New methods to model what the whales are communicating about will guide us when we are making progress in the translation task or, alternately, rule out proposed translations.” The sound sperm whales use to communicate and navigate via echolocation is filled with enigmas, as is any unknown form of communication. Therefore, we are uncertain not only what the codas mean in the sperm whale communication system but also how we can test and confirm our assumptions.
As part of the team’s effort, new advances in artificial intelligence will be employed to address these questions. Using AI models, Begus develops speech recognition systems that learn a language like children do: without supervision, texts, imitation, and fantasy. In addition, he is testing whether models that learn human language from speech can also learn the communication system of sperm whales with his team in the Berkeley Speech and Computation Lab. In addition to underwater microphones, drones, and robotic fish for silently tracking and recording sperm whale communication, CETI collaborators from more than a dozen institutions worldwide are also working hard on other project areas.
Prof. David Gruber, the CETI’s leader, teaches biology and environmental science at Baruch College. A fellowship year at Radcliffe College brought Gruber, Goldwasser, and Michael Bronstein together in 2018. The CETI goal was formulated during a year-long seminar that explored many possible applications of machine learning in science.
Read More: Berkeley.
Join the discussion and participate in awesome giveaways in our mobile Telegram group. Join Curiosmos on Telegram Today. t.me/Curiosmos
