A new model gives wildlife managers the ability to collect the data they need within minutes
When states want to measure quail populations, the process can be grueling, time consuming and expensive.
It means listening to phone calls for hours in the field. Or leave a recording device in the field to pick up what sounds are being made, then listen to that audio hours later. Then repeat this process until there is enough information to make population estimates.
But a new model developed by researchers at the University of Georgia aims to streamline this process. By using artificial intelligence to analyze terabytes of quail call recordings, the process gives wildlife managers the ability to collect the data they need in minutes.
“The model is very accurate and records between 80% and 100% of all calls, even in the noisiest recordings. So you could make a recording, send it through our model and it will tell you how many quail sounds the recorder has heard,” said James Martin, an associate professor at the UGA Warnell School of Forestry and Natural Resources, who has been working on the project for about five years, in conjunction with the Georgia Department of Natural Resources.
“With this new model, you can analyze terabytes of data in seconds, and that allows us to scale up monitoring so you can literally place hundreds of these devices and cover a much larger area with much less effort than in the past.”
The software represents approximately five years of work by Martin, postdoctoral researcher Victoria Nolan, and numerous key contributors who worked with a codewriter to create the model. It’s also part of a larger shift taking place in wildlife research, where computer algorithms are now helping with work that once took humans thousands of hours to complete.
Computers are getting smarter at, for example, recognizing specific sounds or certain properties in photos and sound recordings. For researchers like Martin, this means that hours once spent on tasks like listening to audio or watching video from game cameras can now be done by a computer, freeing up valuable time to focus on other aspects of a project.
The new tool could also be a valuable resource for state and federal agencies seeking information about their quail populations, but with limited resources to devote to a project. “So I think this is something states could do to replace their current monitoring with acoustic recording devices,” Martin added.
The success of the software was recently documented in the magazine of Remote Sensing in Ecology and Conservation.
As the software becomes more used and exposed to sounds from new geographies, Martin said, it gets even “smarter.” As it is, quail offer different types of calls. But when the software is exposed to a variety of sounds that aren’t quail, he said, it’s better able to distinguish the right calls from the ambient sounds of the grasses and trees around it.
Over time, the software will become more critical.
“So that’s why you have to keep giving it training data, and when you move geographic areas, you come across new sounds that you haven’t trained the model for,” he added. “It’s always about adjustment.”
AI better than humans at detecting blue whale calls
Victoria Nolan et al, The development of a convolutional neural network for the automatic detection of Northern Bobwhite Colinus virginianus covey calls, Remote Sensing in Ecology and Conservation (2022). DOI: 10.1002/rse2.294
Quote: AI answers the call for quail information (2022, October 18) retrieved October 19, 2022 from https://phys.org/news/2022-10-ai-quail.html
This document is copyrighted. Other than fair dealing for personal study or research, nothing may be reproduced without written permission. The content is provided for informational purposes only.