A group of international scientists launched a new ambitious project to listen to, contextualize, translate and interpret the communication of Sperm Whales. This goal was to “talk” to these magnificent marine animals.
The initiative, called Project CETI (Cetacean Translation Initiative), is harnessing the power of artificial intelligence to interpret clicking sounds, or ‘codas,’ Sperm whales make to communicate with one another.
Researchers are using natural language processing or NLP – a subfield of artificial intelligence focused on processing written and spoken human language – which will be trained four billion Sperm whale codas.
According to the researchers, the plan is to have the AI match each sound with a context. This will take at least five more years.
If the team achieves its goals, the next step is to create and deploy an interactive chatbot that can engage in dialogue with Sperm Whales living in the wild.
Michael Bronstein, the lead of machine learning for Project CETI, told Hakai: ‘If we discover that there is an entire civilization basically under our nose — maybe it will result in some shift in the way that we treat our environment.
“And maybe it’ll result in more respect and appreciation for the living world.
Scroll down for video

The goal is to have the AI match whale sounds with a context. This will take at least five more years according to the researchers. The next step is to create and deploy an interactive chatbot that can interact with Sperm whales in the wild if the team achieves its goals.
The brain of the Sperm Whale is five times more than that of a human, and it is also considered to be one of the largest on Earth.
Despite being different brains, both whales as well as humans have complex communication systems and live in family groups.
‘These [mammals]If they are with other whales, they will make a clicking noise at different frequencies. Is this a code or a real language? Professor Dan Tchernov of University of Haifa’s Leon H. Charney School of Marine Sciences, who is one of its leaders, spoke to Israel 21c.
‘Right now our database isn’t comprehensive enough to know the answer.

Project CETI (Cetacean Translation Initiative), a project that uses artificial intelligence to interpret clicking sounds or ‘codas’ Sperm whales make in order to communicate with one other, is called Project CETI.
“However, machine learning and advanced linguistics have allowed us to realize that if we gather enough data about the voices of our subjects, their context in which they are used and understood, their behavior and motivation behind the sounds, we can develop an algorithm that will determine if they have an authentic language.
CETI has laid out its five-year plan, which includes deploying dozens of state-of-the-art devices to listen to sperm whale communication, interpret their voices, and then attempt to communicate back,’ Harvard John A. Paulson School of Engineering and Applied Science, which is part of Project CETI, shared in a statement.
“The interdisciplinary team will need to develop the most delicate robotic technology to date to listen and contextualize whale sounds; design and construct a hydrophone array for studying a population of whales; build a big data pipeline to examine the recorded data, decode it using advanced machine-learning and natural language processing; and finally, create a big data pipeline to analyze the data and decode it using data science.
The project, which began in 2020, included experts from Harvard and MIT, Imperial College London (City University of New York), University of Haifa, UC Berkeley and the Institute for Scientific Interchange.
Researchers have already started to create non-invasive soft devices to attach and collect sounds from the whales.

Researchers have already begun their work on developing non-invasive soft device to attach to whales and collect sounds.

The tags must adhere to the skin of each whale without causing damage. They can withstand speeds of up to 30 mph, depths as high as 6,561 feet, and temperatures as low at a few degrees Fahrenheit.

A prototype, which uses suction cups inspired from octopus or squid, is currently being tested. Cups that can securely but gently latch onto the whale have been developed by such designs.
The tags will be unique in that must stick to the whale’s skin without causing damage and have to withstand speeds up to 30 miles per hour, depths up to 6,561 feet and temperatures as low as a few degrees Fahrenheit.
A prototype is currently being tests, that uses suction cups inspired by octopus and squid – the whale’s natural prey.
These cups can be used to gently, but firmly latch onto the whale.
Although the lab testing was promising, the next step will be to ship the devices to whales out in the open ocean.
The prototype contains up to three hydrophones which can provide context about the directionality between whales.

After the team has collected its four billion codas the NLP will develop a system analogous with human language models that generates grammatically accurate whale utterances. These will be uploaded to the chatbot for the open ocean
Once the team collects its four billion codas, the NLP will then develop ‘a system analogous to human language models that generates grammatically correct whale utterances’ that will be uploaded to the chatbot set for the open ocean.
The CETI researchers acknowledge that their search for meanings in whale codas may not have produced anything of interest, but they know the feat is well worth the effort.
Hakai Magazine: David Gruber, a City University of New York marine biologist and program leader, stated that one of the greatest risks to our species is that they could be boring.
“But we don’t believe this to be the case. Biologists have always found that animals are a fascinating subject matter when they look at it closely.