Researchers Use AI to Translate Whale Communication

A team of researchers at Texas A&M University is currently taking advantage of recent advancements in the field of artificial intelligence (AI) in order to translate a rather unusual form of communication: sperm whale vocalizations. The researchers have attained a large corpus of data on the different clicks and vocalizations that the whales make in order to communicate with each other, and believe that the improvements in AI technology and machine learning over the last couple of years could allow us to develop translation systems that allow humans to understand the whales’ complex communication system.

The research team includes members from Texas A&M’s Geochemical and Environmental Research Group (GERG) as well as researchers from Project CETI, the Cetacean Translation Initiative. Project CETI consists of a team of researchers in the field of linguistics, marine biology, and technology, and seeks to uncover information on the communication systems of non-human animals by investigating sperm whales, which have been documented to have one of the most complex animal communication systems in the animal kingdom.


Recently, Project CETI hosted a workshop at UC Berkeley, where researchers discussed the ongoing work they’ve been doing to decode animal communication. At the workshop, Project CETI’s head of machine learning presented some of the next steps the researchers will take in analyzing and decoding the acoustic data encoded in the whales’ vocalizations.

The sperm whale is a particularly apt animal for such research, as it boasts the largest brain in the animal kingdom. Evolutionary anthropologists and linguists have long been interested in animal communication, as these more rudimentary systems — things like whale vocalizations or birdsong — can provide some insight into how language evolved and developed in humans. Now, AI specialists are throwing their hat into the ring in an effort to develop translation systems that will allow us to understand some of the most complex non-human communication documented to date. “GERG and Texas A&M are excited to be part of this project and we were pleased to be approached to build these systems,” said Anthony Knap, Texas A&M’s principal investigator for the project.

Andrew Warner
Andrew Warner is a writer from Sacramento. He received his B.A. in linguistics and English from UCLA and is currently working toward an M.A. in applied linguistics at Columbia University. His writing has been published in Language Magazine, Sactown Magazine, and The Takeout.

Related Articles

Weekly Digest

Subscribe to stay updated