In the 1970s, a gorilla named Koko gained worldwide attention for her ability to use human sign language, as trainers were able to teach her American Sign Language and she communicated with humans in more than a thousand sign languages. But skeptics say cacao and other learning animals like chimpanzees and dolphins don't understand what they're referring to, and that attempts to teach human language to other animal species are already doomed.
Although a group of scientists believes that understanding how these creatures communicate with each other is more useful than trying to find out whether animals can communicate in human language.
Among those scientists, Karen Packer, a researcher at the University of British Columbia in Canada and a member of the Harvard-Radcliffe Institute for Advanced Study, said: “We need to understand non-human communication on our own terms, just as scientists are now doing it using advanced sensors and AI technologies to monitor leaves. “Communicate and learn about different organisms, including: plants.”
The field of digital bioacoustics is the subject of Bakker's new book, The Sound of Life: How Digital Technology Can Bring Us Closer to the World of Plants and Animals.
In an interview with Scientific American, Karen Bakker discusses how technology helps people communicate with creatures like bats and bees, and how these conversations are forcing us to rethink our relationships with other species. It also tells the story of man's attempts to communicate with animals.
“In the middle of the 20th century, there were many attempts to teach non-human species, especially primates, to prove that an animal has language, we must show that it can learn human language and travel through time; this is a human-centered point.”
Packer thinks it's best to think about the ability of these planet-sharing creatures to communicate in complex ways in their own way. "For example, we can't expect bees to speak human language, but these insects do indeed convey interesting language based on vibration, movement and location," she said.
This is where the development of a less anthropocentric science helps us. As a science (digital bioacoustics) - the study of sounds made by living things - digital bioacoustics is on the rise, revealing fascinating insights into how organisms communicate with each other.
She added that this scientific approach does not ask: Can animals talk like humans? Instead, a larger question arises: Can animals communicate complex information to each other, and how do they do it?
Today's research takes a very different approach, starting with recording the sounds made by animals and even plants, and then using artificial intelligence and machine learning techniques to analyze large amounts of data to find patterns and match them with behaviors to determine if there is such a complex message conveyed by the sounds of these creatures.
What researchers are doing today is not trying to teach these species human language, but building a dictionary of the signals and sounds that these species make and then trying to figure out what those signals and those sounds mean to the species.
How do researchers do it?
Computational bioacoustics relies on the use of very small, portable, and lightweight digital recorders, such as micromicrophones, which scientists are installing in a variety of field locations, from the Arctic to the Amazon jungle, where they can be mounted on turtle shells or on the ground. They can be placed at the bottom of the sea or attached to birds, and they can continuously record the sounds of these creatures, 24 hours a day, 7 days a week, in remote areas that scientists cannot easily reach.
What researchers are doing today is not trying to teach these species human language, but building a dictionary of the signals and sounds that these species make and then trying to figure out what those signals and those sounds mean to the species.
How do researchers do it?
Computational bioacoustics relies on the use of very small, portable, and lightweight digital recorders, such as micromicrophones, which scientists are installing in a variety of field locations, from the Arctic to the Amazon jungle, where they can be mounted on turtle shells or on the ground. They can be placed at the bottom of the sea or attached to birds, and they can continuously record the sounds of these creatures, 24 hours a day, 7 days a week, in remote areas that scientists cannot easily reach.
These devices record large amounts of data, and this is where AI comes in, because the same natural language processing algorithms we use in tools like translation can also be used to recognize patterns in the way non-humans communicate.
How can AI analyze these communication patterns?
In his book, Parker discusses the research of researcher Yossi Yovel, who spent two and a half months observing and recording nearly twenty species of Egyptian fruit bats, then his team used artificial intelligence software to analyze 15,000 audio clips. And the algorithm associates specific sounds with interactions in specific social scenarios captured on video, for example b- when bats swarm for food.
In this study, the team was able to categorize most of the sounds made by bats and concluded that bats use a more complex conversational language than previously thought because bats argue about food, distinguish between genders when communicating, and even have their own name.
Mother bats speak to their young in a motherly tone, human mothers raise their voices to guide their offspring, while mother bats use a low voice to teach their offspring certain skills, causing offspring to elicit atypical responses, or to learn certain words or puck specific signals as they grow. suggest that bats participate in vocal learning.
It's a great example of how deep learning can infer these patterns from the data collected by all these sensors and microphones and reveal things we wouldn't have access to without it.
Since most bats communicate via ultrasound and these waves are above the range of our hearing and they speak much faster than we can, in order to hear them we have to slow them down and lower the frequency, we can't use artificial intelligence and machine learning technologies that generate specific patterns and be able to use those patterns to communicate with colonies of bats Or influencing cell behavior is what researchers are currently doing.
How do researchers talk to bees?
In another chapter of the book, Parker discusses the work of a well-known researcher (Tim Landgraf) as Faisal observes bees and the dynamics they use to communicate with each other.
Now, she says, advances in AI technology through computer vision and natural language processing have enabled the team to develop algorithms that can track specific bees and determine how those bees' signals affect the behavior of the rest of the bees. Beehive.
The researchers found that the bees emit semantic signals such as stop, move, and pay attention, in addition to other, more complex signals that express distress and direct group orders.
The researchers (Tim Landgraf) were also able to create a bee-sized robot named RoboBee and outfit it with all the communication symbols that can be extracted from the bee's language. After testing seven or eight robot prototypes, he created a robotic bee that could enter the hive and order members of the hive to respond to their commands.
Robotic bees can give commands to stop or move, and even enable some complex forms of communication between bees, such as the circle dance that bees typically perform to lead the rest of the hive to where the nectar is.
finally:
According to Parker, this scientific method and newer technologies such as artificial intelligence raise many philosophical and ethical questions. You might envision using these techniques to protect bees from harm by directing them to areas that are safe for nectar rather than those infested with high levels of pesticides.
You can also imagine using wildlife to harm animals by trying to domesticate them or to control behaviors that affect the planet's ecosystems.
Ideas about non-human stages of development and complex communication also raise very important philosophical questions. Is human language unique to other creatures?