Artificial intelligence reveals a hidden secret in the language of black sciences

In a new scientific discovery that could change our way of understanding and monitoring wildlife, researchers from the British University of Exeter have succeeded in discovering a “secret” and new type of lion roar that was not known to scientists before, using advanced artificial intelligence techniques.

For many years, zoologists believed that lions had a single, distinct roar pattern, which they used to communicate and establish dominance. But the study The new report, published in the journal Ecology and Evolution, proved that lions possess a “second type” of roar, which scientists described as the “medium roar.”

This hidden sound appears alongside the well-known traditional roar, and was not easy to distinguish by the human ear alone or by traditional methods, which explains why it has remained unknown all these decades.

Machine accuracy

The research team developed artificial intelligence algorithms to analyze audio recordings of lions with extreme accuracy. Instead of relying on the human ear, which might be mistaken or biased, the software deconstructed the lions’ “sound signature” and categorized it automatically.

The result was significant, as the system achieved a classification accuracy of 95.4%, and demonstrated that the lion roar sequence clearly contains these two types (full roar and medium roar).

The study’s lead researcher, Jonathan Grocott, from the University of Exeter, said in a statement, a copy of which was obtained by Al Jazeera Net: “Lions’ roars are not just a symbol, but rather unique fingerprints that can be used to estimate population sizes and monitor individual animals.”

“Until now, identifying this roar has relied heavily on expert estimates, which may introduce human bias. Our new approach using artificial intelligence promises more accurate and less subjective monitoring, which is critical for conservationists working to protect dwindling lion populations.”

Preserving the kings of the jungle

On the other hand, traditional methods for studying lion communities (such as tracking tracks or cameras) may be difficult and expensive in forests and large reserves, but using “passive acoustic monitoring” in the presence of this discovery will make tracking lions easier and more accurate.

Thanks to this precise classification, scientists can now distinguish individual lions from each other based solely on their voices, which helps in counting their numbers more accurately without disturbing them.

With the number of wild lions in Africa declining to only between 20,000 and 25,000 (a halving in the last 25 years), any technological tool that helps monitor and protect them is considered a lifeline.