From chirps to melodies, birdsong exhibits a range as varied as a musical composition.
However, ovenbirds — stout, ground-nesting warblers — possess a very limited song selection.
The male ovenbird has a singular tune: a quick melody often referred to as teacher-teacher-teacher.
To the average listener, all ovenbirds may appear identical in their singing, but each bird has its unique rendition.
Currently, biologists are employing AI technology to monitor the wellbeing of ovenbird populations through their distinctive songs.
Research released this month in the journal Nature Methods utilized AI to recognize individual ovenbirds from recordings collected in Alberta and Pennsylvania.
Focused on one tune
The ovenbird’s “particular” song selection made it an ideal subject for this research, according to lead researcher Sam Lapp, a postdoctoral fellow at the University of Colorado.
“They are quite dedicated to singing a very specific song,” Lapp stated during an interview with CBC News.
“Many birds develop a vocal signature over their lifetime, often incorporating several different songs into what we refer to as their repertoire.
“In contrast, the ovenbird uniquely maintains a stable song over time, performing only one version of it.”
Ovenbirds perform their primary song in the spring to establish their territory and attract mates, and this refrain persists throughout their lives.
Under Lapp’s leadership during his tenure with the Department of Biological Sciences at the University of Pittsburgh, the research team utilized recordings from nearly 130 locations in Pennsylvania collected over four years.
They trained an AI model to identify unique song features and group similar songs together, based on the assumption that each bird would return to its familiar forest area each spring.
The team analyzed 9,072 hours of audio to monitor 405 individual birds. The researchers then validated the model using an additional set of recordings gathered from the mixed wood and boreal forests of Alberta during the 2016 spring breeding season.
According to Lapp, the model could connect songs to individual birds with an accuracy rate of up to 96 percent.
This method is significantly less labor-intensive than traditional leg banding, where biologists capture birds, attach markers to their legs, and then attempt to recapture them the following spring.
The research idea arose from a partnership with Alberta biologists looking to locate birds across extensive distances by utilizing a grid of microphones.
Astonishing discovery
Lapp noted that human ears are unable to differentiate individual ovenbirds in real time due to the rapid nature of their vocal variations.
However, when viewed on spectrographs — visual representations of sound — each bird’s unique song becomes evident, Lapp explained.
“I started to wonder if my eyes could recognize these patterns, could we train a machine to identify them and perform automated individual recognition?”
“Even for those deeply interested in birdwatching and species identification, it’s surprising and thrilling to realize that this level of detail exists, which we might have overlooked … but is indeed present if you listen closely enough.”
Erin Bayne, a biology professor at the University of Alberta and a co-author of the study, has been listening to ovenbirds for many years and is familiar with the challenges of annual population tracking.
During his PhD, Bayne dedicated thousands of hours to monitoring banded birds, to the point where he could distinguish individuals by their unique vocal differences.
“What that bird is saying over there differs slightly from what the other one is saying,” he remarked. “I was able to identify who was who.
“They are incredibly courteous singers. It’s amusing, really. One bird will sing, and then the other will wait until it’s finished before immediately responding.”
Birds communicate with each other
Bayne stated that the findings of this study reveal that bird communication is significantly more intricate than most people realize, and that conversations occur even if we do not comprehend the language.
Just as humans exhibit distinct speech patterns, biologists are discovering how particular animals develop their unique vocal identities, according to Bayne.
“It’s not surprising that individual birds or animals have detectable differences. I believe that as we continue to explore this, it will become increasingly evident.”
Although the model has its limitations, the researchers aspire to offer a non-invasive method for assessing songbird health simply by listening.
The AI system struggles to accurately track the various calls of female ovenbirds or the basic sounds made by young birds.
The current system would also be inadequate for monitoring species like magpies or mockingbirds that frequently mimic new sounds or have numerous songs.
It would be most effective for birds that maintain a consistent song throughout their lives — approximately 30 percent of songbird species — along with specific species of bats and owls.
The researchers have made the program code available on open-source platforms, enabling other biologists to utilize it for tracking the survival of different songbird populations or even other species.
Richard Hedley, a statistical ecologist with the Alberta Biodiversity Monitoring Institute who did not participate in the study, noted its remarkable potential.
While AI acoustic monitoring has been employed for years to track birds at the species level, demonstrating that it can differentiate between individual birds is a remarkable achievement, he stated.
There is a vast amount of existing acoustic recordings that could be utilized in innovative manners, he added.
“Catching a single bird can take countless hours. Thus, the concept of monitoring them simply by placing a microphone in their territory is very promising.”
