BirdyHunt : bird recognition application based on the sound

• Main Author: Li, Wei-Ting
• Other Authors: Owen Noel Newton Fernando
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2021
• Subjects: Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence
• Online Access: https://hdl.handle.net/10356/153231
• Institution: Nanyang Technological University

This project aims to create a simple Android application that can recognize the species of the birds through their calls. The app will be used in ways similar to popular music recognition apps. The user will record the bird sounds heard and the app will return with the name, picture, and short introduction on the species. Different approaches have been used over the years to tackle this problem including decision trees, nearest neighbor, and deep learning. This project will focus on the use of deep learning techniques and the use of ResNeSt, a Split-Attention Networks published in a paper in 2020. The interface of the app is inspired by popular music recognition applications such as Shazam and Musixmatch to ensure the ease of use of the application. The application was formally known as AnimalHunt but rebranded after limiting the scope to birds only.