From chirps to melodies, the variety in birdsong resembles that of a grand orchestra.
However, ovenbirds — stout, terrestrial warblers — possess a very restricted collection of songs.
The male ovenbird has just one song: a quick tune often referred to as teacher-teacher-teacher.
While to many, all ovenbirds may seem identical in their singing, each one has its own unique rendition.
Currently, biologists are harnessing artificial intelligence to monitor the well-being of ovenbird populations based solely on their distinctive songs.
A study released this month in Nature Methods utilized AI technology to identify individual ovenbirds within audio recordings collected from Alberta and Pennsylvania.
Single-mindedness
The ovenbird’s “specific” song selection made it an ideal candidate for this research, according to lead researcher Sam Lapp, a postdoctoral scholar at the University of Colorado.
“They tend to stick to one particular song,” Lapp explained in a CBC News interview.
“Many birds develop a vocal signature throughout their lives, often incorporating various songs into what we refer to as their repertoire.
“In contrast, the ovenbird features a stable song that remains unchanged over time and only performs one version of it.”
Ovenbirds typically sing their primary song during spring to assert their territory and attract a partner, and this song continues to echo throughout their lives.
Under Lapp’s guidance at the University of Pittsburgh’s Department of Biological Sciences, the research team analyzed recordings gathered from nearly 130 locations in Pennsylvania over a four-year period.
They trained an AI model to identify unique features in the songs and group similar songs together, based on the assumption that each bird would return to the same forest area every spring.
The researchers processed an impressive 9,072 hours of audio to monitor 405 individual ovenbirds. They then validated the model using another set of recordings obtained from the mixed wood and boreal forests of Alberta during the spring breeding season in 2016.
According to Lapp, the model can associate songs with 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, carefully mark their legs, and then try to catch them again the next spring.
Lapp noted that the idea for this research stemmed from a collaboration with biologists in Alberta who were working on locating birds over large areas using a network of microphones.
Astonishing Discovery
Lapp explained that human ears struggle to differentiate individual ovenbirds in real time because their vocal subtleties occur too rapidly.
However, through spectrographs—visual depictions of sound—each bird’s unique song becomes clearer, according to Lapp.
“I began to wonder, if I can see these patterns with my eyes, can we teach a machine to recognize these patterns and automate the process of individual recognition?”
“Even for those who are passionate about birds and keen on species identification, it’s quite remarkable and thrilling to realize that this depth of information exists that we’ve been overlooking… yet is available if you listen closely enough.”
Erin Bayne, a biology professor at the University of Alberta and co-author of the study, has been listening to ovenbirds for many years and understands the challenges of annual population monitoring.
During his doctoral studies, Bayne dedicated countless hours to tracking banded birds, to the extent that he could distinguish individuals based on their distinctive vocal differences.
“What that bird is saying over there differs slightly from that bird over there,” he noted. “I recognized who was who.
“They sing with remarkable politeness. It’s amusing, really. One bird will sing, then another will wait for it to finish, but then will immediately sing right after.”
Birds Communicate
Bayne stated that the research indicates that avian communication is much more intricate than most people realize, and that discussions are taking place, even if we do not grasp their language.
Similar to how humans have unique speech patterns, biologists are discovering that certain animals also cultivate their own voices, Bayne remarked.
“It’s not surprising that individual birds or animals have detectable differences. I believe that the more we explore this, the more evident it will become.”
Although the model has its limitations, the study’s authors aspire for it to provide a less intrusive means of assessing the health of songbird populations through mere listening.
The AI system struggles to accurately track the various calls of female ovenbirds or the basic sounds produced by young birds.
This current program would also be inadequate for monitoring species like magpies and mockingbirds, which constantly imitate new sounds or have extensive song repertoires.
It is most effective for birds that consistently sing one unchanging song throughout their lives — approximately 30 percent of songbird species — along with certain bat and owl species.
The researchers have made the program code available on open-source platforms to assist other biologists in monitoring the survival of different songbird populations or even various species.
Richard Hedley, a statistical ecologist at the Alberta Biodiversity Monitoring Institute who did not participate in the study, remarked on its significant potential.
While AI acoustic monitoring has been utilized for years to track birds at the species level, demonstrating its ability to differentiate individual birds is a remarkable achievement, he stated.
There exists a vast array of acoustic recordings that could be utilized in innovative ways, he added.
“Catching a single bird can take many hours. Thus, the idea that we can achieve this by placing a microphone in the bird’s territory is incredibly promising.”
Tag: rewrite-pending