Dog or elephant translators: can technology improve human-animal communication?

Of the more than 8 million species on the planet, humans only understand the language of one of them.

After decades of searching for ways to communicate with animals, several scientists have turned to artificial intelligence to detect patterns in the sounds they make and in their behavior, understand what they mean, and try to interact with them.

Despite the promising advances of multiple investigations, creating translators of elephants, dogs or whales poses multiple challenges.

Eva Meijer, author of the book

explains that animals "talk all the time, to each other and in multispecies environments to survive, make friends, discuss social norms, flirt, and have a good life."

"There is scientific evidence that they have complex languages, cultures and internal lives, and that they fall in love with and mourn their partners," says the expert.

As he recounts in his book, the dolphins call each other by their names;

the prairie dogs describe the intruders in great detail;

bats love to gossip and grammatical structures can be found in the songs of some birds.

While wild chimpanzees understand each other with dozens of different gestures, bees dance to communicate and are able to recognize and remember human faces.

Studying the language and behavior of animals "is important not only to find out how they communicate with each other, but also to investigate how they communicate with us."

Some like dogs, birds and horses are capable of learning words.

For example, a border collie can memorize more than 1,000, according to a study published in the journal

.

In addition, there are animals that respond to the tone of voice and body language, according to Melody: "Soft tones indicate friendship, while hard or strong tones can be threatening."

Touch can be used "as a reward with dogs and horses."

Artificial intelligence to 'talk' to animals

Multiple scientists have turned to artificial intelligence and other technologies to understand and improve this communication.

"Sensors can help us record, analyze and interpret many different animal signals, even those that might be difficult for humans' limited sensory apparatus to detect," says Clara Mancini, a researcher in animal-computer interaction at the Open University in United Kingdom.

The promoters of the Wild Dolphin Project have compiled for more than 30 years a database of dolphin behavior and the sounds they make, which are three: whistles for long-distance communication and between mothers and calves when they are separated, a kind of clicks to navigate and navigate and the so-called burst pulses, which are usually several widely spaced clicks that they use to socialize and fight when they are close.

The objective of this project is to create machine learning algorithms to discover patterns in these sounds and develop systems capable of generating "words" to be able to interact with dolphins in wild environments.

There are many similar projects.

Elephant Voices researchers have created an online ethogram of elephant vocalizations and behaviors in Kenya and Mozambique.

In it, they describe, for example, that when these animals come out of the water after playing, they usually make sounds similar to a trumpet.

Another team of researchers has developed software to automatically detect, analyze and categorize the ultrasonic vocalizations of rodents.

It is called DeepSqueak and it has also been used on lemurs, whales and other marine animals.

While some scientists have developed systems to detect when chickens are making distress calls, others are trying to understand dogs.

“We have conducted studies of dog whines recorded with microphones on their collars and used machine learning to determine if the whining was sad, because it misses its owner, for example, or happy, because it anticipates a play session,” says Melody Jackson, professor at the Georgia Institute of Technology and an expert on dog-computer interaction.

The challenges of creating “translators”

Although there are researchers who have identified the structure and part of the meaning of the vocalizations of some animals, creating "translators" entails multiple challenges.

To begin with, understanding the semantic and emotional meaning of what they communicate is a very complex task, as Mancini argues: "We are not in their mind and we do not have the same physical, sensory and cognitive characteristics through which they experience the world."

Failure to take these differences and complexities into account can carry a risk: “Trivializing what animals communicate and getting it wrong”.

Added to this is the fact that current technologies require environmental or portable sensors that are not always practical.

This is indicated by Jackson, who points out that the appropriate cameras are sometimes not available and that filming an animal in motion well enough for video analysis is very difficult.

Also, interpreting their communication based solely on vocal expression leaves out other channels that may be important in understanding what they mean.

This is the case of behavior, according to Mancini.

Animals also communicate with their actions, their gestures and even their facial expressions.

For example, if two groups of elephants come together and fold their ears while rapidly waving them, they express a warm greeting that is part of their welcoming ceremonies, according to the Elephant Voices ethogram.

On the other hand, sheep's facial expressions can indicate that they are in pain.

In fact, computer scientists from the University of Cambridge have developed an artificial intelligence system to analyze their faces and detect when they suffer damage.

Some researchers study dogs' postures and behaviors to predict how they're feeling, and sometimes turn to biometrics to try to pinpoint changes in heart rate, breathing and temperature that could give clues to their emotions, Jackson says.

While some of these “canine interpretation” systems use body sensors to measure body position and movement, others use cameras to record and analyze the videos.

Vests for training robotic dogs and bees

Being able to communicate with animals can be useful in multiple contexts.

Jackson's team, for example, has developed technology that allows a human handler to guide a search-and-rescue dog remotely using vibrating motors attached to a vest.

"We've also created portable computers that allow a service dog to call emergency services with a GPS location if its owner is having a seizure," she says.

Humans may never be able to sing like a whale or buzz like a bee, but perhaps machines can.

In fact, a team of German researchers has made a biomimetic robot called RoboBee that imitates the dances used by bees to communicate.

The results have been successful: with this robot, they claim to have managed to recruit real bees and guide them to specific destinations.

The advances are very promising, but it is still too early to predict whether animal translators will ever exist.

Jackson believes that as computers and sensors become smaller and more capable, tiny implantable systems will be developed that provide more clues about their behavior and how to achieve two-way communication.

“Our understanding of animal vocalizations, movements, and gestures should one day be automatable, and technology could give us the means to mimic their behaviors, perhaps virtually, and allow us to communicate concepts to them,” he concludes.

You can follow

Subscribe to continue reading

Read without limits

Keep reading

I'm already a subscriber