From chirps to melodies, avian sounds can be as varied as an orchestra.
However, ovenbirds — stout, terrestrial warblers — possess a very limited song selection.
The male ovenbird has a singular tune: a brisk melody often referred to as teacher-teacher-teacher.
To the untrained ear, all ovenbirds may appear identical, but each individual has its own unique rendition.
Currently, biologists are employing artificial intelligence to monitor the health of ovenbird populations based solely on their distinctive songs.
A study released this month in Nature Methods utilized AI to differentiate individual ovenbirds from audio recordings gathered in Alberta and Pennsylvania.
Focused on one song
The ovenbird’s “particular” song selection made it an ideal candidate for this research, explained lead researcher Sam Lapp, a postdoctoral fellow at the University of Colorado.
“They are quite consistent in singing one very specific melody,” Lapp stated during an interview with CBC News.
“Many birds can develop a vocal signature over their lifetime which often includes various songs as part of their repertoire.
“In contrast, the ovenbird stands out by maintaining one stable song throughout its life and only performing that one version.”
Ovenbirds use their primary song during the spring to defend their territory and attract a mate, and this song remains a constant throughout their lives.
Under Lapp’s leadership at the Department of Biological Sciences at the University of Pittsburgh, the research team analyzed recordings collected from nearly 130 sites in Pennsylvania over a span of four years.
They trained an AI model to identify features that render songs distinct and group similar songs together, operating under the assumption that each bird would return to its specific forest area each spring.
The team examined 9,072 hours of audio to monitor 405 individual birds. Subsequently, they tested the model against another collection of recordings from the mixed wood and boreal forests of Alberta during the 2016 spring breeding season.
Lapp noted that the model could associate songs with individual birds with an accuracy of up to 96 percent.
This method is significantly less labor-intensive compared to traditional leg banding, where biologists capture birds, attach markers to their legs, and then aim to recapture them the following spring.
He mentioned that the research idea stemmed from a partnership with biologists in Alberta who were trying to locate birds over large areas using a network of microphones.
Astonishing discoveries
According to Lapp, human ears struggle to differentiate individual ovenbirds in real-time due to the swift nature of their vocal nuances.
However, when viewed on spectrographs — visual displays of sound — each bird’s unique song becomes apparent, he explained.
“I began to wonder, if my eyes can detect these patterns, can we train a machine to recognize these patterns and automate individual identification?”
“Even for those who are passionate about birdwatching and species identification, it’s astonishing and exciting, almost magical, that this wealth of information exists that we have simply overlooked… but it is indeed there if you listen closely enough.”
Erin Bayne, a biology professor at the University of Alberta and co-author of the study, has been monitoring ovenbirds for many years and understands the challenges of annual population assessments.
During his doctoral studies, Bayne dedicated tens of thousands of hours to tracking banded birds, to the extent that he could distinguish individuals based on their unique vocal differences.
“What one bird is saying over there is a bit different from what the other bird is saying,” he remarked. “I could identify who was who.”
“Additionally, they are remarkably courteous while singing. It’s rather amusing, actually. One bird will sing, then wait for the other to finish before immediately singing back.”
Birds communicating
Bayne pointed out that this study reveals that bird communication is significantly more intricate than most people realize, indicating that conversations are occurring, even if we don’t grasp the language.
Just as humans have unique speech patterns, biologists are discovering that certain animals also develop their own distinctive voices, Bayne noted.
“It is not surprising that individual birds or animals possess detectable differences. The more we investigate, the more evident it will become that this is indeed true.”
Although the model has its limitations, the study’s researchers aspire to offer a less invasive method of understanding the health of songbird populations, simply by listening.
The AI system currently struggles to accurately track the various calls of female ovenbirds or the simpler sounds produced by young birds.
This approach may also be inadequate for monitoring species like magpies and mockingbirds, which frequently mimic new sounds or sing a variety of different songs.
It is expected to perform best with birds that maintain a single, consistent song throughout their lives — roughly 30 percent of songbird species — as well as certain bats and owls.
The researchers have made the program code available on open-source platforms, aiming to assist other biologists in tracking the survival of various songbird populations or even different species.
Richard Hedley, a statistical ecologist at the Alberta Biodiversity Monitoring Institute who was not part of the research, stated that this work holds remarkable potential.
While AI acoustic monitoring has been employed to track birds at the species level for years, demonstrating the capability to distinguish between individual birds is an impressive accomplishment, he remarked.
There is a vast amount of existing acoustic data that could be utilized in innovative ways, he added.
“Catching a single bird can take many hours. Therefore, the notion that we can achieve this by simply placing a microphone in the bird’s territory is highly promising.”
Tag: rewrite-pending
