Certificate of Completion
THIS ACKNOWLEDGES THAT
HAS COMPLETED THE MAY-SUMMER 2024 DATA SCIENCE BOOT CAMP
Andrew Merwin
Roman Holowinsky, PhD
JUNE 10, 2024
DIRECTOR
DATE
TEAM
Chirp Checker
Andrew Merwin, Caleb Fong, B Mede, Yang Yang, Robert Cass, Calvin Yost-Wolff
The nocturnal soundscapes of late summer and autumn are replete with the familiar chirps, trills, and buzzes of singing insects. But these cryptic performers often remain anonymous and underappreciated.
The goal of this project was to build machine learning models to identify the presence of insects in sound files and to coarsely categorize the sounds as crickets, katydids, or cicadas.
Both Support Vector Classifiers and Convolutional Neural Networks were able to identify insects songs to the broad categories of cricket, katydid, and cicada with 90% accuracy or higher.
In the future, similar, more sophisticated models could be applied to filtering large volumes of passively recorded audio from ecological studies of insects and could power apps that identify insect songs to the species level.