Latest news with #DolphinGemma


Mint
29-04-2025
- Science
- Mint
Clicks & growls: Why AI's hearing the call of the wild
Google has used artificial intelligence (AI) to decode and mimic dolphin sounds, advancing our understanding of marine life. But can AI truly outperform human insight in interpreting animal communication? Also Read | Return of Indian tech brands: Is it for real this time? Dolphins are famously socially skilled, intelligent, agile, joyful and playful, thus sharing many emotional similarities with (some) humans. Just as British ethologist Jane Goodall studied the social and family interactions of wild chimpanzees, Denise Herzing has studied dolphin communication in the wild since 1985, making The Wild Dolphin Project (WDP) the longest running underwater dolphin research project in the world. Google, in partnership with Georgia Tech and WDP, used an AI model to analyse vocal patterns much like a language model, identifying structure and predicting sequences. Also Read | Why Pakistan's trade ban is more sound than fury Dolphins use distinct sounds for different social situations: whistles act like names for individual identification, especially between mothers and calves; squawks often occur during conflicts; and click buzzes are used in courtship. DolphinGemma, Google's 400 million parameter model that runs on Pixel6 smartphones, decodes these sounds by combining audio tech with data from WDP acquired by studying wild dolphins in the Bahamas. On National Dolphin Day (14 April), Google showcased advances to its AI model that can now analyse dolphin vocal patterns and generate realistic, dolphin-like sounds. Also Read | Return of the dire wolf: Is this a Game of Clones? AI is being used to detect how parrots, crows, wolves, whales, chimpanzees and octopuses communicate. NatureLM-audio is the first audio-language model built for animal sounds and can generalize to unseen species. Other projects use AI and robotics to decode sperm whale clicks, or listen to elephants to detect possible warning calls and mating signals. It aids conservation by monitoring endangered species. Decoding communication reveals ecosystem health, alerting us to pollution and climate change. It enhances human-animal interactions and fosters empathy. AI, combined with satellite imagery, camera traps and bioacoustics, is being used in Guacamaya to monitor deforestation and protect the Amazon, a collaboration between Universidad de los Andes, Instituto SINCHI, Instituto Humboldt, Planet Labs and Microsoft AI for Good Lab. AI can detect animal sound patterns, but without context— is the animal mating, feeding or in danger?—its interpretations are limited. The risk of humans assuming animals 'talk" like humans do, messy field data, and species-specific behaviours can complicate analysis. AI might identify correlations but not true meaning or intent. Human intuition helps. These systems often require custom models and extensive resources, making large-scale, accurate decoding of animal communication a complex effort.
Yahoo
27-04-2025
- Science
- Yahoo
Google working to decode dolphin communication using AI
Cracking the dolphin code. Dolphins are one of the smartest animals on Earth and have been revered for thousands of years for their intelligence, emotions and social interaction with humans. Now Google is using artificial intelligence (AI) to try and understand how they communicate with one another – with the hope that one day humans could use the technology to chat with the friendly finned mammals. Google has teamed up with researchers at Georgia Institute of Technology and the Wild Dolphin Project (WDP), a Florida-based non-profit which has been studying and recording dolphin sounds for 40 years, to build the new AI model called DolphinGemma. 'Super Pod' Of More Than 1,500 Dolphins Off California Coast Captured On Drone Video For decades, WDP has correlated sound types with behavioral contexts. For instance, signature whistles have been used by mothers and calves to reunite, while burst pulse "squawks" are often observed during dolphin fights, researchers said, according to a Google blog on the project. Read On The Fox News App Click "buzzes" are often used during courtship or chasing sharks. Now, using the vast data gathered by WDP, Google has built DolphinGemma, building upon Google's own AI lightweight open model, known as Gemma. DolphinGemma has been trained to analyze the vast library of recordings to detect patterns, structures, and even potential "meanings" behind the dolphin communications or vocalizations. Over time, DolphinGemma will try to organize the dolphin sounds into categories — almost like words, sentences, or expressions in human language. Chinese Humanoid Robot With Eagle-eye Vision And Powerful Ai "By identifying recurring sound patterns, clusters and reliable sequences, the model can help researchers uncover hidden structures and potential meanings within the dolphins' natural communication — a task previously requiring immense human effort," a post on Google about the project reads. "Eventually, these patterns, augmented with synthetic sounds created by the researchers to refer to objects with which the dolphins like to play, may establish a shared vocabulary with the dolphins for interactive communication." DolphinGemma uses Google's Pixel phone technology, specifically the audio recording technology used in Pixel devices, to make clean, high-quality sound recordings of dolphin vocalizations. The Pixel phone technology can separate out dolphin clicks and whistles from background noise like waves, boat engines, or underwater static. That clean audio is critical for AI models like DolphinGemma, because messy, noisy data would confuse the AI, researchers said. Google says it plans to release DolphinGemma as an open model this summer, allowing researchers around the world to use and adapt it. Although it's trained on Atlantic spotted dolphins, the model could also help study other species like bottlenose or spinner dolphins with some fine-tuning, researchers said. "By providing tools like DolphinGemma, we hope to give researchers worldwide the tools to mine their own acoustic datasets, accelerate the search for patterns and collectively deepen our understanding of these intelligent marine mammals," the blog post article source: Google working to decode dolphin communication using AI


Fox News
27-04-2025
- Science
- Fox News
Google working to decode dolphin communication using AI
Cracking the dolphin code. Dolphins are one of the smartest animals on Earth and have been revered for thousands of years for their intelligence, emotions and social interaction with humans. Now Google is using artificial intelligence (AI) to try and understand how they communicate with one another – with the hope that one day humans could use the technology to chat with the friendly finned mammals. Google has teamed up with researchers at Georgia Institute of Technology and the Wild Dolphin Project (WDP), a Florida-based non-profit which has been studying and recording dolphin sounds for 40 years, to build the new AI model called DolphinGemma. 'SUPER POD' OF MORE THAN 1,500 DOLPHINS OFF CALIFORNIA COAST CAPTURED ON DRONE VIDEO For decades, WDP has correlated sound types with behavioral contexts. For instance, signature whistles have been used by mothers and calves to reunite, while burst pulse "squawks" are often observed during dolphin fights, researchers said, according to a Google blog on the project. Click "buzzes" are often used during courtship or chasing sharks. Now, using the vast data gathered by WDP, Google has built DolphinGemma, building upon Google's own AI lightweight open model, known as Gemma. DolphinGemma has been trained to analyze the vast library of recordings to detect patterns, structures, and even potential "meanings" behind the dolphin communications or vocalizations. Over time, DolphinGemma will try to organize the dolphin sounds into categories — almost like words, sentences, or expressions in human language. CHINESE HUMANOID ROBOT WITH EAGLE-EYE VISION AND POWERFUL AI "By identifying recurring sound patterns, clusters and reliable sequences, the model can help researchers uncover hidden structures and potential meanings within the dolphins' natural communication — a task previously requiring immense human effort," a post on Google about the project reads. "Eventually, these patterns, augmented with synthetic sounds created by the researchers to refer to objects with which the dolphins like to play, may establish a shared vocabulary with the dolphins for interactive communication." DolphinGemma uses Google's Pixel phone technology, specifically the audio recording technology used in Pixel devices, to make clean, high-quality sound recordings of dolphin vocalizations. The Pixel phone technology can separate out dolphin clicks and whistles from background noise like waves, boat engines, or underwater static. That clean audio is critical for AI models like DolphinGemma, because messy, noisy data would confuse the AI, researchers said. Google says it plans to release DolphinGemma as an open model this summer, allowing researchers around the world to use and adapt it. Although it's trained on Atlantic spotted dolphins, the model could also help study other species like bottlenose or spinner dolphins with some fine-tuning, researchers said. "By providing tools like DolphinGemma, we hope to give researchers worldwide the tools to mine their own acoustic datasets, accelerate the search for patterns and collectively deepen our understanding of these intelligent marine mammals," the blog post reads.
Yahoo
24-04-2025
- Science
- Yahoo
Study Finds Dolphins Can Make Human Vowel Sounds as Experts Launch AI-Powered Dolphin Translation
New research reveals that dolphins can mimic human vowel sounds, indicating their potential for interspecies communication The study highlights a dolphin named Zeus and the marine mammal's intentional communication with humans Google's AI model, DolphinGemma, is advancing human-dolphin communication, helping researchers identify mimicked sounds and better respond to dolphins' requests Could humans have entire conversations with dolphins in the future? A new study found that the marine mammals already know part of the human language. In a press announcement published on Tuesday, April 23, J & J Publishing, a dolphin-focused education and science publisher, released a research paper titled "Novel Dolphin Vocalization." It reveals the "intricate cognitive capabilities of dolphins and their sophisticated understanding of mimicry as a tool for interspecies communication." The study notes that dolphins have developed the ability to produce specific sounds, measured by complexity, frequency, amplitude, and duration, that mimic vocalizations resembling the human vowel sounds, "A, E, O, and U." Related: Once-Extinct Bird Is Now Laying Eggs in the Wild for the First Time in Nearly 40 Years Unlike humans, who auditorily communicate with their mouths and vocal cords, dolphins' vocalizations come from their blowholes and air sacs. The sounds emitted can create a range of octaves, some even too complex for the human ear. Researchers from SpeakDolphin focused on a dolphin named Zeus for the study after noticing the animal's spontaneous vocalizations. "I missed it at first," confessed lead researcher Jack Kassewitz. "Because they live in aquatic environments, dolphins' brains have adapted to process sound much faster than humans. As researchers, it's often challenging to recognize dolphins' very fast-paced acoustic patterns while recording on location." However, once the researcher caught on, he realized the vowel sounds were distinct from Zeus' normal dolphin vocalization. "I am convinced that some dolphins, Zeus in particular, are as determined to communicate with us as we are with them," said Kassewitz. "We have heard anecdotal reports over the decades of similar dolphin mimicry. Finally, we can analyze these specialized vocalizations with our advanced software." Related: 'Extremely Rare' Hybrid 'Rockaroni' Penguins Caught on Camera Evolving 'Before Our Eyes' (Exclusive) Kassewitz explained the difference between the sounds dolphins make connecting with each other versus the sounds made towards humans above water. "When dolphins vocalize above the water surface, it is almost certainly to get the attention of nearby humans," he noted. "Zeus' vowel vocalizations — made with his head above water and directed at our research team — demonstrated how intentional his efforts were." The study comes only weeks after Google announced its progress in human-dolphin communication using AI. On April 14, in celebration of National Dolphin Day, Google — in collaboration with researchers at Georgia Tech (led by Google DeepMind Research Scientist and Georgia Tech professor Dr. Thad Starner) and field researchers led by the Research Director and founder of the Wild Dolphin Project (WDP) Dr. Denise Herzing — revealed the developments made to their foundational AI model, DolphinGemma. The AI model uses the SoundStream tokenizer, a Google audio technology, to recreate dolphin sounds to identify recurring sound patterns, clusters, and reliable sequences. Related: New Details Emerge About the Seizure and Death of Pet Influencer Peanut the Squirrel From decades of underwater observations, WDP has analyzed specific sounds with behavioral correlations for dolphins. For example, signature whistles (or names) are used to call the calves back to them; burst-pulse 'squawks' are used during fights; and clicking 'buzzes' are heard during courting and shark chases. WDP has furthered its communication exploration by creating the Cetacean Hearing Augmentation Telemetry (or CHAT) system, alongside Georgia Institute of Technology, to establish a more stable, more straightforward shared vocabulary with the marine mammals. Researchers use the CHAT system to create synthetic dolphin sounds and then associate those sounds with objects that the dolphins they're studying enjoy, including sargassum, seagrass, or scarves. From there, the CHAT system listens for the dolphins to mimic the sounds it creates, identifies which whistle was mimicked, and informs the researcher which object the dolphin is referring to. This allows the researcher to provide the requested item quickly, building the vocabulary and connection between the two. With DolphinGemma, CHAT can anticipate and identify mimicked sounds faster. This increases the speed at which researchers can respond to the dolphin's requests, allowing for a more fluid communication sequence. Read the original article on People
Yahoo
15-04-2025
- Science
- Yahoo
Google's newest AI model is designed to help study dolphin 'speech'
Google's AI research lab, Google DeepMind, says that it has created an AI model that can help decipher dolphin vocalizations, supporting research efforts to better understand how dolphins communicate. The model, called DolphinGemma, was trained using data from the Wild Dolphin Project (WDP), a nonprofit that studies Atlantic spotted dolphins and their behaviors. Built on Google's open Gemma series of models, DolphinGemma, which can generate "dolphin-like" sound sequences, is efficient enough to run on phones, Google says. This summer, WDP plans to use Google's Pixel 9 smartphone to power a platform that can create synthetic dolphin vocalizations and listen to dolphin sounds for a matching "reply." WDP previously was using the Pixel 6 to conduct this work, Google says, and upgrading to the Pixel 9 will enable researchers at the organization to run AI models and template-matching algorithms at the same time, according to Google.