Researchers at New York University (NYU) and the Cornell Lab of Ornithology are helping to track the nighttime migratory patterns of birds by teaching a computer to recognize their flight calls.
The researchers used a technique called acoustic monitoring in conjunction with algorithms to gather better information than they had in the past.
"The intro point to go from human to computer is about thinking of these sounds in terms of frequency and time, and figuring out how to measure that in increasing detail and feed that information into the machine's listening models," says Cornell researcher Andrew Farnsworth.
The system includes sensors equipped with a spectral template detector that scans the audio as it comes in, checking for potential matches. "When a potential match is identified, it snaps roughly one second of that audio centered around the detection and sends that through the server," says NYU's Justin Salamon.
The researchers "teach" the system to recognize flight calls by providing it with a large collection of recordings, and then they use "unsupervised feature learning," in which the program builds a statistical model of the specific patterns that are represented of a certain species. The goal is to be able to put names to nocturnally-migrating species in an automated way in real time.
From Public Radio International (MN)
View Full Article
Abstracts Copyright © 2016 Information Inc., Bethesda, Maryland, USA
No entries found