How artificial intelligence helps us decipher animal languages

For years, artificial intelligence (AI) and machine learning have been used to analyze and translate human languages. This area of ​​research has helped expand channels of human communication and has led to new language-based technologies such as advanced chatbots and intelligent voice command devices.

Today, artificial intelligence is being used to study animal communication, as researchers aim to decipher the animals’ languages ​​to support conservation and sustainability efforts.

said Katie Zakarian, CEO and co-founder of the Earth Species Project (ESP), an organization that uses artificial intelligence to decode animal communications. “With this progress, we expect that we are moving rapidly toward a world in which two-way communication with other species is likely.”

ESP, a non-profit organization based in California, US, has been developing machine learning systems that can decode animal communication by identifying patterns in behavioral ecology research for years. This includes analyzing large datasets containing visual, oral, and physical animal communication. The goal, the researchers say, is to determine the circumstances under which an animal emits a communication signal, how the receiving animal reacts, and which signals are relevant in influencing actions.

“It’s not as simple as Dr. Doolittle’s device.”

Karen Packer, professor at the University of British Columbia

AI analysis is possible and, in fact, necessary today, given the vast amounts of animal communication data being collected using increasingly sophisticated sensors and recording devices, according to experts. This includes devices that can be set up and record sound in previously inaccessible habitats such as the deep sea and mountain tops. ESP, for example, works with more than 40 biologists and environmental partner institutions around the world to source datasets.

How artificial intelligence analyzes animal communication

said Karen Packer, a professor at the University of British Columbia and author of the book Sounds of Life: How digital technology is bringing us closer to the animal and plant worlds. “But this only takes us so far. In order to really understand animal communication, we have to relate that to behavior.”

The AI-powered analysis of animal communication includes datasets for both bioacoustics, recording individual organisms, and environmental acoustics, recording entire ecosystems, according to the experts. In October 2022, ESP published the first publicly available benchmark for measuring the performance of machine learning algorithms in bioacoustics research. The system – known as BEANS (Balancing Animal Sounds) – uses 10 datasets from different animal communication and lays a foundation for machine learning classification and detection performance.

Data sets being studied under various efforts to decode animal communication include recordings from a range of species such as birds, amphibians, monkeys, elephants, and insects such as honey bees. Communication from domesticated cats and dogs is also being studied. However, experts note that communication between cetaceans — whales, dolphins, and other marine mammals — is particularly promising.

“Whales are particularly interesting because of their long history — 34 million years as a cultural, socially learned species,” Zakarian explained. “And because—because light does not diffuse well underwater—more of their communication is forced through the vocal tract.”

Researchers assert that bioacoustics and AI-powered analysis of animal communication can greatly enhance environmental research and conservation efforts.

For example, in 2021, researchers used audio recordings to identify a new species of blue whale in the Indian Ocean. The research team explained nature Detailed article on the discovery.

Moreover, listening to ecosystems and decoding animal communications can help ecologists gauge the health of the natural environment, experts say. This includes, for example, developing a better understanding of how distributive human activity such as noise or logging affects animal populations. In Costa Rica, for example, sound recordings have recently been used to assess the development and health of reforested areas of the rainforest.

“By observing sounds that come from nature, we can look for changes in social structure, transmission of cultural information or physiological stress,” Zakarian said.

AI analysis of animal communication has also been used to help establish marine animal protection zones. For example, off the West Coast of the United States, researchers have used artificial intelligence to analyze marine communication recordings as well as shipping route data to create “mobile marine protected areas” and predict potential alliances between animals and ships.

“Understanding what animals are saying is the first step to giving other species on the planet a ‘voice’ in conversations about our environment,” said Kay Firth Butterfield, Head of Artificial Intelligence and Machine Learning at the World Economic Forum. “For example, should the whales be told to dive away by boat when that fundamentally changes their feeding or should the boats change course?”

Smarter, Faster: The Big Think newsletter

Subscribe to get unexpected, surprising, and poignant stories delivered to your inbox every Thursday

There are ethical concerns that researchers face, too. This includes, in particular, the possibility of causing harm by creating two-way communication channels between humans and animals – or animals and machines.

“We’re not entirely sure what effect it will have on the animals and whether they want to engage in some of the conversations,” Packer said. “Maybe if they talk to us, they’ll tell us to leave.”

Researchers are taking steps to address and mitigate concerns about animal harm and exploitation. For example, ESP works with its partners to develop a set of principles to guide its research and to ensure that it always supports animal conservation and animal welfare.

“We’re not sure yet what all the real-world applications of this technology are,” Zakarian stated. “But we believe unlocking an understanding of other species’ connections will be very important as we work to change the way humans see our role, and as we figure out how to co-exist on this planet.”

This article has been reprinted with permission from the World Economic Forum, where it was originally published.

Leave a Comment