{"id":942733,"date":"2026-10-23T01:18:14","date_gmt":"2026-10-23T01:18:14","guid":{"rendered":"https:\/\/www.newsbeep.com\/ca\/942733\/"},"modified":"2026-10-23T01:18:14","modified_gmt":"2026-10-23T01:18:14","slug":"scientists-use-ai-to-figure-out-how-to-speak-whale","status":"publish","type":"post","link":"https:\/\/www.newsbeep.com\/ca\/942733\/","title":{"rendered":"Scientists Use AI to Figure Out How to Speak Whale"},"content":{"rendered":"<p class=\"wp-block-paragraph\">LONDON \u2013 Arriving on the Caribbean island of Dominica for the first time \u201cfeels a little bit like stepping into Jurassic Park,\u201d said Giovanni Petri, professor in the Network Science Institute at Northeastern University London.<\/p>\n<p class=\"wp-block-paragraph\">Known as Nature Island, Dominica comprises thick rainforest, volcanic mountains with dramatic waterfalls, and vibrant coral reefs that fall away into deep blue waters. But when Petri came here in August 2022, he didn\u2019t come to study the extinct Tyrannosaurus Rex but the largest predator that lives on our planet today\u2014the sperm whale.<\/p>\n<p class=\"wp-block-paragraph\">These massive marine mammals that can grow up to 60 feet and weigh as much as 57 tons can be found across the world\u2019s oceans, from the equator to the frigid oceans around the two poles. But Dominica is one of only a handful of places they live throughout the year, offering the chance to observe their activity and listen in on their conversations.<\/p>\n<p class=\"wp-block-paragraph\">This matters because Petri is one of the lead researchers for Project CETI, a non-profit scientific initiative that aims to use advanced artificial intelligence, machine learning and whale-friendly robotics to record and ultimately decode the acoustic communication of sperm whales.<\/p>\n<p class=\"wp-block-paragraph\">Founded in 2020, Project CETI is the largest effort at interspecies communication in history, comprising more than 50 researchers at leading research institutions in the U.S and worldwide. In Dominica, its researchers have built a comprehensive research platform, where they observe whales, engineer listening devices and develop new ways to analyze data.<\/p>\n<p class=\"wp-block-paragraph\">\u201cWe have three boats, drones carrying tags, hydrophones in the water and whole pipelines of analysis trying to find structure in the data,\u201d said Petri.<\/p>\n<p class=\"wp-block-paragraph\">But how does it actually work?<\/p>\n<p class=\"wp-block-paragraph\">Petri explained to Northeastern Global News how the enigmatic whale sounds  can be turned into understandable language.<\/p>\n<p><a data-scroll-scene-nav-link=\"\" data-target=\"scroll-scene-observe\" href=\"#scroll-scene-observe\">Observe<\/a><a data-scroll-scene-nav-link=\"\" data-target=\"scroll-scene-listen\" href=\"#scroll-scene-listen\">Listen<\/a><a data-scroll-scene-nav-link=\"\" data-target=\"scroll-scene-annotate\" href=\"#scroll-scene-annotate\">Annotate<\/a><a data-scroll-scene-nav-link=\"\" data-target=\"scroll-scene-analyze\" href=\"#scroll-scene-analyze\">Analyze<\/a><a data-scroll-scene-nav-link=\"\" data-target=\"scroll-scene-patterns\" href=\"#scroll-scene-patterns\">Patterns<\/a><a data-scroll-scene-nav-link=\"\" data-target=\"scroll-scene-close\" href=\"#scroll-scene-close\">What lies ahead<\/a><\/p>\n<p class=\"wp-block-paragraph\">Each day, a sperm whale needs to eat 3 to 4% of its body weight to survive \u2013 and for a species that can weigh up to 57 tons, this means a staggering quantity of food. By some estimates, a fully grown sperm whale can eat more than a ton of food in 24 hours.<\/p>\n<p class=\"wp-block-paragraph\">Dominica\u2019s year-round community offers the chance to observe exactly how they manage this. Project CETI uses boats to gather unique video and has developed a method for attaching complex sensors to the whales\u2019 gigantic bodies. They fly quadcopter drones that stick advanced wildlife tracking devices directly on to the side of the whale. These biologger devices attach with non-harmful suckers and contain GPS trackers with depth sensors, movement trackers, heart rate monitors and more.<\/p>\n<p class=\"wp-block-paragraph\">The resident whales embark on deep \u201cforaging dives\u201d, spending 45 minutes to an hour, descending from the warm sunlit waters around Dominica into the extreme pressure and near-complete darkness of around 700 metres depth, explained Petri.<\/p>\n<p class=\"wp-block-paragraph\">\u201cWhen they resurface, they spend about 15, 20 minutes breathing, re-oxygenating after long dives,\u201d he said. This is the basic routine played out, again and again, with some whales synchronizing their deep dives.<\/p>\n<p class=\"is-family-sans is-weight-light is-size-16 has-line-height-24 has-letter-spacing-comfortable expanded:is-family-sans expanded:is-weight-light expanded:is-size-19 expanded:has-line-height-28 expanded:has-letter-spacing-normal extra-large:is-size-23 extra-large:has-line-height-36 has-on-surface-variant-color\">Bonus trivia: Sperm whales can dive for over a kilometer below the surface to feast on deep sea creatures such as giant squid, with some estimates based on their stomach contents claiming they could dive as deep as 3 kilometers, almost two miles.<\/p>\n<p class=\"wp-block-paragraph\">In Dominica, Project Ceti employs underwater microphones known as hydrophone, and is developing semi-autonomous marine \u201cgliders\u201d which could follow whales for weeks without a human driver. These together can detect sperm whales\u2019 vocalizations. They show that they emit a rhythmic series of short clicks that researchers call codas. Researchers believe these codas are \u201clike little words,\u201d said Petri.<\/p>\n<p class=\"wp-block-paragraph\">\u201cMost of the socialisation that we know of happens on the surface,\u201d explained Petri. \u201cSo when they come up, often what you see is that animals that are a few hundred metres apart start drifting toward each other, and then they start coda-producing rapidly once they are very close, within one body length of each other. Then, they start chatting a lot,\u201d he said.<\/p>\n<p class=\"wp-block-paragraph\">These regular interactions give rise to a set of questions that are driving current research: are they coordinating their synchronised dives? Are some whales leading and others following? And what other information might they be passing on?<\/p>\n<p class=\"is-family-sans is-weight-light is-size-16 has-line-height-24 has-letter-spacing-comfortable expanded:is-family-sans expanded:is-weight-light expanded:is-size-19 expanded:has-line-height-28 expanded:has-letter-spacing-normal extra-large:is-size-23 extra-large:has-line-height-36 has-on-surface-variant-color\">Bonus trivia: The Dominica site aims to be almost totally autonomous and aims to soon feature observational drones taking off from solar powered platforms, deploying tags, and streaming data back to a central database \u2013 no human intervention required.<\/p>\n<p class=\"wp-block-paragraph\">In Dominica, all of this data is streamed \u2013 from biologgers, hydrophones, and gliders \u2013 into a massive \u201cdata vault,\u201d explained Petri. Then comes one of the hardest bits \u2013 annotating the acoustic data so that it can be analyzed by data scientists.<\/p>\n<p class=\"wp-block-paragraph\">First, the system will detect clicks and identify whether they are the kinds of regular pinging clicks sperm whales use in the dark, deep ocean for echolocation, or the rhythmic clicks they emit when socialising.<\/p>\n<p class=\"wp-block-paragraph\">\u201cThen you want to sort of figure out which clicks go together in one coda,\u201d Petri said. This can be like trying to listen into a busy party, he said \u201cbecause you might have multiple animals speaking at the same time, or you might have noise, so it might be that some clicks are missing, and so you have to infer.\u201d<\/p>\n<p class=\"wp-block-paragraph\">A process of source-separation sorts which of the resident whales \u2013 which the team have named \u2013 is talking.<\/p>\n<p class=\"wp-block-paragraph\">\u201cSo you want to say, well, this coda comes from Pinchy,\u201d one of the named whales, he said.\u201dThis other coda is actually Digit talking.\u201d<\/p>\n<p class=\"wp-block-paragraph\">Ultimately this aims to translate sound into usable data. A large part of this is automated using AI pattern-recognition systems, and the team is working on making it more effective, \u201cbut we also still have humans in the loop\u201d, including the Project CETI biology team who are trained to help curate and check that the automatic annotation actually makes sense,\u201d Petri said.<\/p>\n<p class=\"is-family-sans is-weight-light is-size-16 has-line-height-24 has-letter-spacing-comfortable expanded:is-family-sans expanded:is-weight-light expanded:is-size-19 expanded:has-line-height-28 expanded:has-letter-spacing-normal extra-large:is-size-23 extra-large:has-line-height-36 has-on-surface-variant-color\">Bonus trivia: Project CETI researchers have developed WhaleLM, a large language model s similar to those powering ChatGPT to analyze and predict the communication patterns of sperm whales<\/p>\n<p class=\"wp-block-paragraph\">\u201cNow, once you have the clean annotated acoustic data, then you can pair it with behavioural data, for example, the accelerometers, the depth sensors, and so on and on,\u201d said Petri.<\/p>\n<p class=\"wp-block-paragraph\">This has been a driving ambition of Project CETI\u2019s work in Dominica, to develop sustained monitoring that can pair sound and action. Together, they hope these can provide something like a Rosetta Stone for whale language, showing which repeated codas correlate with key events.<\/p>\n<p class=\"wp-block-paragraph\">Already, simply by feeding recordings of whale chat into machine learning models, the Project CETI team has found that around eight to ten codas can predict whether they are about to head off on deep foraging dives or not, explains Pertri.<\/p>\n<p class=\"wp-block-paragraph\">But this is far from cracking whale language. The CETI team aims to understand the basic units of language. The current working hypothesis is that they have a basic set of symbols \u2013 a sort of fundamental phonetic alphabet \u2013 that have been identified using neural networks, which extract measurable features from the sound. By examining the timing, rhythm, spectral shape of codas, they have identified two vowel-like sounds and recent research has shown that they may modulate these, like tones in Mandarin, to express different senses. But their meaning is not yet known, Petri said.<\/p>\n<p>Pinchy\u2019s \u201ca\u201d vowel<\/p>\n<p>Pinchy\u2019s \u201ci\u201d vowel<br \/>\nProject CETI recordings<\/p>\n<p class=\"is-family-sans is-weight-light is-size-16 has-line-height-24 has-letter-spacing-comfortable expanded:is-family-sans expanded:is-weight-light expanded:is-size-19 expanded:has-line-height-28 expanded:has-letter-spacing-normal extra-large:is-size-23 extra-large:has-line-height-36 has-on-surface-variant-color\">Bonus trivia: Sperm whales have distinct dialects that mark membership of social clans. Researchers studying whales off Dominica found that families from two different clans shared the same waters, but only socialised with families that used their own dialect.<\/p>\n<p class=\"wp-block-paragraph\">Petri is a theoretical physicist who wrote his thesis at CERN, the European Organization for Nuclear Research, where scientists explore the fundamental structure of the Universe. But he was invited to join Project CETI at the outset, as an expert in topological data analysis, which looks at data points as a connected network and uses mathematical shapes to find hidden patterns in complex information.<\/p>\n<p class=\"wp-block-paragraph\">His team uses statistical, network and machine-learning tools to identify recurring patterns and relationships, which can help ultimately shed light on the meaning of the codas.<\/p>\n<p class=\"wp-block-paragraph\">\u201cAI is fundamental for us, from coordinating flying fleets of drones to finding what the regularities are in the language that they have,\u201d said Petri.<\/p>\n<p class=\"wp-block-paragraph\">In a previous paper in the journal Science, Petri used the same kinds of network science tools that his colleagues at Northeastern NetSI institutes use to trace how viral infections spread through hospitals or misinformation spreads through social interaction. This showed for the first time how two different family units cooperated to help a newborn calf to breathe in its first hours after birth, while it is still becoming buoyant and learning to swim. This analysis allowed them to see which whales each individual is paying attention to, with attention shifting from the mother before the birth, to the calf post-partum.<\/p>\n<p class=\"wp-block-paragraph\">To understand more about foraging dives, they aim to use these systems to find out if certain influential whales are initiating deep dives, and others are following.<\/p>\n<p class=\"wp-block-paragraph\">\u201cIs there a clear leadership within this community \u2013 are older females typically the ones beginning dives or not? Are the calves the ones that are driving communication by basically going around and chatting with everyone?\u201d Petri said.<\/p>\n<p class=\"wp-block-paragraph\">From here, it should become clear whether there is a particular coda \u2013 a whale word \u2013 that triggers a dive.<\/p>\n<p class=\"is-family-sans is-weight-light is-size-16 has-line-height-24 has-letter-spacing-comfortable expanded:is-family-sans expanded:is-weight-light expanded:is-size-19 expanded:has-line-height-28 expanded:has-letter-spacing-normal extra-large:is-size-23 extra-large:has-line-height-36 has-on-surface-variant-color\">Bonus trivia: Young sperm whales appear to \u201cbabble\u201d, producing a wider variety of codas than adults. It takes several years for calves to consistently reproduce the patterns used by their family.<\/p>\n<p class=\"wp-block-paragraph\">Petri recalls two whales diving together in perfect sync, with biologgers attached, allowing the team to record \u201cboth acoustically and behaviorally\u201d how this coordination took place. Over time, another whale, named Peach, drifted closer and joined the yo-yoing process of feeding and resting, explained Petri.<\/p>\n<p class=\"wp-block-paragraph\">\u201cSo for the last few dives, there are actually the three of them diving together going up and down in very close proximity,\u201d he said.<\/p>\n<p class=\"wp-block-paragraph\">Petri\u2019s team is using proximity networks and acoustic analysis to investigate how it all came together. They aim to understand if echolocation plays a role.<\/p>\n<p class=\"wp-block-paragraph\">As sperm whales dive into the ink black depths, their eyesight fails and they use clicks to sense squid. In the dark, Project CETI has hypothesized they may eavesdrop on each other\u2019s clicks, to sense where large groups of squid are \u2013 like two people at night in a forest who are \u201cusing the reflection of each other\u2019s torch lights to find their way in the dark,\u201d said Petri.<\/p>\n<p class=\"wp-block-paragraph\">If whales do use one another\u2019s clicks to find prey, the sounds could serve simultaneously as both communication and perception.<\/p>\n<p class=\"wp-block-paragraph\">\u201cSuch complicated events probe the limits of what we can ever hope to translate,\u201d Petri said.<\/p>\n<p>Audio, data and visuals<\/p>\n<p>Audio: Project CETI. Recordings of the whale named Pinchy from the Project CETI press kit, re-encoded for the web with no other processing.<\/p>\n<p>Click timings were detected automatically from the same Project CETI recording; they are not hand-annotated.<\/p>\n<p>Coda rhythms and the whale-to-whale exchange: Dominica sperm whale coda dataset (Sharma et al., Nature Communications, 2024), <a href=\"https:\/\/doi.org\/10.5281\/zenodo.10817697\" target=\"_blank\" rel=\"noreferrer noopener nofollow\">doi:10.5281\/zenodo.10817697<\/a>, CC BY 4.0.<\/p>\n<p>Visualizations are schematic illustrations of the research process. The 3D whales and equipment are stylized models. Nothing on this page shows what a coda means.<\/p>\n<p class=\"wp-block-paragraph\">Interactive design and development by Zach Christensen<\/p>\n","protected":false},"excerpt":{"rendered":"LONDON \u2013 Arriving on the Caribbean island of Dominica for the first time \u201cfeels a little bit like&hellip;\n","protected":false},"author":2,"featured_media":942734,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[7],"tags":[2837,276,49,48,2523,796,189640,21231,66],"class_list":["post-942733","post","type-post","status-publish","format-standard","has-post-thumbnail","category-science","tag-animals","tag-artificial-intelligence","tag-ca","tag-canada","tag-language","tag-machine-learning","tag-network-science","tag-ocean","tag-science"],"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/posts\/942733","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/comments?post=942733"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/posts\/942733\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/media\/942734"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/media?parent=942733"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/categories?post=942733"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/tags?post=942733"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}